perf: avoid rebuilding dictionary keys during group emission by Rich-T-kid · Pull Request #27051 · apache/datafusion

Rich-T-kid · 2026-11-03T14:09:07Z

Why

DictionaryGroupValuesColumn stores each group as an index into its deduplicated values. Building the output dictionary currently walks every group again to convert those indices to dictionary keys. With nulls, it also removes the null value with take and remaps the keys.

group slots:  [0, 2, 1, 2]          values: ["a", null, "b"]
                  |                              |
                  | convert + remap              | take out null
                  +---------------+--------------+
                                  |
                            validate every key
                                  |
                                  v
                           DictionaryArray

This makes group emission relatively expensive when Parquet RLE values stay dictionary encoded.

What changed

The group column now stores each inner slot directly in the dictionary's native key type. Nulls use an all-ones placeholder plus a validity bit, so they no longer occupy a slot in the values array. Cached equality compares native keys directly and avoids null checks for non-null batches.

keys:     [0, 1, *, 1] --------+
validity: [1, 1, 0, 1] --------+-- move --> DictionaryArray
values:   ["a", "b"] -----------+
             ^
             +-- keys also identify inner values while grouping

Full emission moves the existing keys, validity, and values directly into the result. Partial emission still compacts the referenced values and remaps the remaining groups. Non-null batches append the validity in one operation after collecting their keys.

On 64-bit platforms an Int32 dictionary with nulls reduces the per-group mapping from an 8-byte usize to a 4-byte key and one validity bit. Without nulls, the validity buffer remains unallocated.

Results

The final rebased head was measured with the existing dictionary_group_values benchmark in three upstream runs: run 1, run 2, and run 3. High-cardinality repeated batches improve by 14%, scalar append/equality by 8–15%, and high-cardinality partial emission is flat to 7% faster. No case regressed consistently across the three runs.

This change is complementary to the dictionary hash and group-mapping reuse in #26826. A current-base stack of both changes was measured in four benchmark groups:

Benchmark	Result	Runs
`dictionary_group_values`	High-cardinality intern and repeated-intern cases are 1.3–1.7x faster, partial emission is 1.5–1.6x faster, and shared-values dictionaries are 1.7–2.7x faster	1, 2, 3
TPC-H	About 1% faster overall with no statistically changed queries	1
TPC-DS	Flat to 5% faster overall, with repeatable 6–23% improvements in Q4, Q11, Q22, Q67, and Q74	1, 2, 3
Partitioned ClickBench	1.7–2.6% faster overall	1, 2, 3

The broad runs enable DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY on both sides.

Rich-T-kid · 2026-11-03T14:09:32Z

run benchmark tpcds

env:
  DATAFUSION_RUNTIME_MEMORY_LIMIT: "32G"
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "32bd81fb226ccf1ae5df67ff340d77baa3e73668"
changed:
  ref: "939ae6c37e6326f2737600e4d60ef2d3673f119a"

adriangbot · 2026-11-03T14:13:25Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6605994558-4304-wlwkd 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing 939ae6c (939ae6c) to 32bd81f diff

Run configuration

run benchmark tpcds
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
  DATAFUSION_RUNTIME_MEMORY_LIMIT: "32G"
baseline:
  ref: "32bd81fb226ccf1ae5df67ff340d77baa3e73668"
changed:
  ref: "939ae6c37e6326f2737600e4d60ef2d3673f119a"

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-11-03T14:24:58Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing 939ae6c (939ae6c) to 32bd81f diff

Run configuration

run benchmark tpcds
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
  DATAFUSION_RUNTIME_MEMORY_LIMIT: "32G"
baseline:
  ref: "32bd81fb226ccf1ae5df67ff340d77baa3e73668"
changed:
  ref: "939ae6c37e6326f2737600e4d60ef2d3673f119a"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and rich-T-kid_dictionary-aggregate-emission
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ rich-T-kid_dictionary-aggregate-emission ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │    5.29 ms │                                  6.05 ms │  1.14x slower │
│ QQuery 2  │   75.29 ms │                                 78.34 ms │     no change │
│ QQuery 3  │   27.99 ms │                                 30.17 ms │  1.08x slower │
│ QQuery 4  │  775.26 ms │                                986.75 ms │  1.27x slower │
│ QQuery 5  │   52.59 ms │                                 53.08 ms │     no change │
│ QQuery 6  │   34.85 ms │                                 35.14 ms │     no change │
│ QQuery 7  │   72.68 ms │                                 73.72 ms │     no change │
│ QQuery 8  │   39.12 ms │                                 39.76 ms │     no change │
│ QQuery 9  │   54.47 ms │                                 51.69 ms │ +1.05x faster │
│ QQuery 10 │   59.82 ms │                                 59.89 ms │     no change │
│ QQuery 11 │  554.67 ms │                                442.07 ms │ +1.25x faster │
│ QQuery 12 │   27.74 ms │                                 24.59 ms │ +1.13x faster │
│ QQuery 13 │  116.71 ms │                                110.79 ms │ +1.05x faster │
│ QQuery 14 │  375.28 ms │                                375.74 ms │     no change │
│ QQuery 15 │   49.70 ms │                                 50.88 ms │     no change │
│ QQuery 16 │    5.96 ms │                                  6.14 ms │     no change │
│ QQuery 17 │   75.88 ms │                                 76.51 ms │     no change │
│ QQuery 18 │  108.38 ms │                                100.71 ms │ +1.08x faster │
│ QQuery 19 │   42.06 ms │                                 40.34 ms │     no change │
│ QQuery 20 │   32.14 ms │                                 30.01 ms │ +1.07x faster │
│ QQuery 21 │   17.56 ms │                                 16.48 ms │ +1.07x faster │
│ QQuery 22 │   71.44 ms │                                 78.20 ms │  1.09x slower │
│ QQuery 23 │  284.39 ms │                                317.53 ms │  1.12x slower │
│ QQuery 24 │  183.15 ms │                                212.43 ms │  1.16x slower │
│ QQuery 25 │  106.46 ms │                                109.89 ms │     no change │
│ QQuery 26 │   45.44 ms │                                 46.38 ms │     no change │
│ QQuery 27 │    6.43 ms │                                  6.84 ms │  1.06x slower │
│ QQuery 28 │   60.21 ms │                                 62.72 ms │     no change │
│ QQuery 29 │   92.70 ms │                                 98.63 ms │  1.06x slower │
│ QQuery 30 │   34.90 ms │                                 38.79 ms │  1.11x slower │
│ QQuery 31 │  107.76 ms │                                115.95 ms │  1.08x slower │
│ QQuery 32 │   19.71 ms │                                 21.89 ms │  1.11x slower │
│ QQuery 33 │   37.29 ms │                                 37.71 ms │     no change │
│ QQuery 34 │    9.08 ms │                                  9.17 ms │     no change │
│ QQuery 35 │   68.48 ms │                                 69.20 ms │     no change │
│ QQuery 36 │    5.56 ms │                                  5.59 ms │     no change │
│ QQuery 37 │    7.20 ms │                                  7.10 ms │     no change │
│ QQuery 38 │   63.31 ms │                                 63.27 ms │     no change │
│ QQuery 39 │   70.70 ms │                                 70.55 ms │     no change │
│ QQuery 40 │   24.43 ms │                                 22.08 ms │ +1.11x faster │
│ QQuery 41 │   14.65 ms │                                 13.55 ms │ +1.08x faster │
│ QQuery 42 │   24.51 ms │                                 23.57 ms │     no change │
│ QQuery 43 │    5.51 ms │                                  5.57 ms │     no change │
│ QQuery 44 │    9.19 ms │                                  9.37 ms │     no change │
│ QQuery 45 │   37.67 ms │                                 37.91 ms │     no change │
│ QQuery 46 │   11.45 ms │                                 11.79 ms │     no change │
│ QQuery 47 │  226.18 ms │                                198.29 ms │ +1.14x faster │
│ QQuery 48 │   88.66 ms │                                 89.41 ms │     no change │
│ QQuery 49 │   68.47 ms │                                 69.09 ms │     no change │
│ QQuery 50 │   56.41 ms │                                 55.95 ms │     no change │
│ QQuery 51 │   92.50 ms │                                 93.58 ms │     no change │
│ QQuery 52 │   24.87 ms │                                 23.27 ms │ +1.07x faster │
│ QQuery 53 │   29.71 ms │                                 28.26 ms │     no change │
│ QQuery 54 │   54.69 ms │                                 51.76 ms │ +1.06x faster │
│ QQuery 55 │   23.09 ms │                                 22.68 ms │     no change │
│ QQuery 56 │   37.81 ms │                                 37.99 ms │     no change │
│ QQuery 57 │  148.99 ms │                                150.68 ms │     no change │
│ QQuery 58 │   99.97 ms │                                104.63 ms │     no change │
│ QQuery 59 │  104.91 ms │                                105.62 ms │     no change │
│ QQuery 60 │   41.64 ms │                                 42.09 ms │     no change │
│ QQuery 61 │    9.03 ms │                                  8.94 ms │     no change │
│ QQuery 62 │   37.04 ms │                                 37.09 ms │     no change │
│ QQuery 63 │   28.26 ms │                                 28.67 ms │     no change │
│ QQuery 64 │  345.50 ms │                                350.59 ms │     no change │
│ QQuery 65 │   74.26 ms │                                 75.15 ms │     no change │
│ QQuery 66 │   70.62 ms │                                 75.98 ms │  1.08x slower │
│ QQuery 67 │  215.99 ms │                                246.45 ms │  1.14x slower │
│ QQuery 68 │   10.89 ms │                                 12.29 ms │  1.13x slower │
│ QQuery 69 │   52.71 ms │                                 55.72 ms │  1.06x slower │
│ QQuery 70 │   95.65 ms │                                 98.15 ms │     no change │
│ QQuery 71 │   33.42 ms │                                 34.23 ms │     no change │
│ QQuery 72 │ 1453.17 ms │                               1580.21 ms │  1.09x slower │
│ QQuery 73 │    8.63 ms │                                  8.83 ms │     no change │
│ QQuery 74 │  234.57 ms │                                238.39 ms │     no change │
│ QQuery 75 │  136.87 ms │                                137.05 ms │     no change │
│ QQuery 76 │   33.71 ms │                                 33.53 ms │     no change │
│ QQuery 77 │   58.01 ms │                                 58.35 ms │     no change │
│ QQuery 78 │  154.32 ms │                                175.89 ms │  1.14x slower │
│ QQuery 79 │   64.93 ms │                                 71.89 ms │  1.11x slower │
│ QQuery 80 │   91.67 ms │                                100.82 ms │  1.10x slower │
│ QQuery 81 │   26.98 ms │                                 30.54 ms │  1.13x slower │
│ QQuery 82 │   16.50 ms │                                 18.33 ms │  1.11x slower │
│ QQuery 83 │   30.66 ms │                                 31.85 ms │     no change │
│ QQuery 84 │   29.76 ms │                                 29.85 ms │     no change │
│ QQuery 85 │   92.91 ms │                                 93.44 ms │     no change │
│ QQuery 86 │   23.44 ms │                                 24.06 ms │     no change │
│ QQuery 87 │   63.20 ms │                                 63.66 ms │     no change │
│ QQuery 88 │   57.32 ms │                                 59.97 ms │     no change │
│ QQuery 89 │   34.61 ms │                                 36.51 ms │  1.05x slower │
│ QQuery 90 │   15.47 ms │                                 15.48 ms │     no change │
│ QQuery 91 │   37.47 ms │                                 37.89 ms │     no change │
│ QQuery 92 │   27.59 ms │                                 27.78 ms │     no change │
│ QQuery 93 │   47.66 ms │                                 47.56 ms │     no change │
│ QQuery 94 │   36.57 ms │                                 36.46 ms │     no change │
│ QQuery 95 │   77.83 ms │                                 77.71 ms │     no change │
│ QQuery 96 │   22.83 ms │                                 23.02 ms │     no change │
│ QQuery 97 │   49.78 ms │                                 49.63 ms │     no change │
│ QQuery 98 │   36.77 ms │                                 37.22 ms │     no change │
│ QQuery 99 │   55.45 ms │                                 55.54 ms │     no change │
└───────────┴────────────┴──────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                       ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                       │ 9095.04ms │
│ Total Time (rich-T-kid_dictionary-aggregate-emission)   │ 9482.58ms │
│ Average Time (HEAD)                                     │   91.87ms │
│ Average Time (rich-T-kid_dictionary-aggregate-emission) │   95.78ms │
│ Queries Faster                                          │        12 │
│ Queries Slower                                          │        22 │
│ Queries with No Change                                  │        65 │
│ Queries with Failure                                    │         0 │
└─────────────────────────────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-T-kid_dictionary-aggregate-emission
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃ rich-T-kid_dictionary-aggregate-emission ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.29 / 5.96 ±1.07 / 8.09 ms │              6.05 / 6.67 ±0.99 / 8.65 ms │  1.12x slower │
│ QQuery 2  │        75.29 / 75.58 ±0.18 / 75.84 ms │           78.34 / 78.99 ±0.41 / 79.61 ms │     no change │
│ QQuery 3  │        27.99 / 28.34 ±0.28 / 28.67 ms │           30.17 / 30.40 ±0.22 / 30.75 ms │  1.07x slower │
│ QQuery 4  │    775.26 / 800.45 ±14.47 / 817.07 ms │       986.75 / 993.60 ±6.72 / 1004.88 ms │  1.24x slower │
│ QQuery 5  │        52.59 / 53.10 ±0.43 / 53.66 ms │           53.08 / 53.46 ±0.28 / 53.82 ms │     no change │
│ QQuery 6  │        34.85 / 34.93 ±0.09 / 35.11 ms │           35.14 / 35.58 ±0.30 / 35.89 ms │     no change │
│ QQuery 7  │        72.68 / 75.70 ±4.15 / 83.84 ms │           73.72 / 75.87 ±3.08 / 81.94 ms │     no change │
│ QQuery 8  │        39.12 / 39.67 ±0.35 / 40.18 ms │           39.76 / 40.09 ±0.37 / 40.78 ms │     no change │
│ QQuery 9  │        54.47 / 55.66 ±1.37 / 58.33 ms │           51.69 / 55.17 ±2.26 / 58.12 ms │     no change │
│ QQuery 10 │        59.82 / 61.81 ±2.19 / 65.97 ms │           59.89 / 62.21 ±2.64 / 67.28 ms │     no change │
│ QQuery 11 │     554.67 / 559.79 ±4.29 / 567.25 ms │       442.07 / 517.70 ±57.64 / 565.66 ms │ +1.08x faster │
│ QQuery 12 │        27.74 / 29.17 ±2.53 / 34.23 ms │           24.59 / 24.84 ±0.31 / 25.44 ms │ +1.17x faster │
│ QQuery 13 │     116.71 / 118.28 ±0.92 / 119.22 ms │        110.79 / 114.12 ±2.84 / 118.59 ms │     no change │
│ QQuery 14 │    375.28 / 387.54 ±10.00 / 400.84 ms │        375.74 / 382.47 ±4.91 / 388.48 ms │     no change │
│ QQuery 15 │        49.70 / 50.78 ±0.79 / 51.94 ms │           50.88 / 51.55 ±0.41 / 51.96 ms │     no change │
│ QQuery 16 │           5.96 / 6.02 ±0.09 / 6.19 ms │              6.14 / 6.33 ±0.20 / 6.72 ms │  1.05x slower │
│ QQuery 17 │        75.88 / 76.67 ±0.94 / 78.39 ms │           76.51 / 77.68 ±0.75 / 78.87 ms │     no change │
│ QQuery 18 │     108.38 / 108.99 ±0.62 / 110.02 ms │        100.71 / 101.49 ±0.89 / 103.18 ms │ +1.07x faster │
│ QQuery 19 │        42.06 / 42.70 ±0.72 / 44.05 ms │           40.34 / 40.80 ±0.33 / 41.22 ms │     no change │
│ QQuery 20 │        32.14 / 32.37 ±0.27 / 32.83 ms │           30.01 / 30.21 ±0.22 / 30.55 ms │ +1.07x faster │
│ QQuery 21 │        17.56 / 17.76 ±0.16 / 17.95 ms │           16.48 / 16.81 ±0.28 / 17.30 ms │ +1.06x faster │
│ QQuery 22 │        71.44 / 73.60 ±2.28 / 77.24 ms │           78.20 / 79.27 ±0.65 / 80.22 ms │  1.08x slower │
│ QQuery 23 │     284.39 / 287.50 ±4.01 / 295.37 ms │        317.53 / 328.14 ±8.17 / 336.27 ms │  1.14x slower │
│ QQuery 24 │     183.15 / 192.44 ±7.83 / 202.18 ms │        212.43 / 214.65 ±1.88 / 217.92 ms │  1.12x slower │
│ QQuery 25 │     106.46 / 108.21 ±1.58 / 111.02 ms │        109.89 / 112.53 ±3.37 / 118.98 ms │     no change │
│ QQuery 26 │        45.44 / 45.80 ±0.40 / 46.52 ms │           46.38 / 46.62 ±0.22 / 46.98 ms │     no change │
│ QQuery 27 │           6.43 / 6.57 ±0.09 / 6.66 ms │              6.84 / 6.89 ±0.05 / 6.97 ms │     no change │
│ QQuery 28 │        60.21 / 61.36 ±1.25 / 63.79 ms │           62.72 / 63.92 ±1.84 / 67.59 ms │     no change │
│ QQuery 29 │        92.70 / 93.83 ±1.75 / 97.30 ms │         98.63 / 100.19 ±1.53 / 103.11 ms │  1.07x slower │
│ QQuery 30 │        34.90 / 35.58 ±0.38 / 36.08 ms │           38.79 / 38.99 ±0.26 / 39.48 ms │  1.10x slower │
│ QQuery 31 │     107.76 / 109.28 ±1.22 / 110.78 ms │        115.95 / 118.70 ±3.56 / 125.35 ms │  1.09x slower │
│ QQuery 32 │        19.71 / 21.08 ±1.82 / 24.67 ms │           21.89 / 22.03 ±0.14 / 22.30 ms │     no change │
│ QQuery 33 │        37.29 / 37.77 ±0.76 / 39.28 ms │           37.71 / 39.18 ±1.38 / 40.93 ms │     no change │
│ QQuery 34 │           9.08 / 9.22 ±0.22 / 9.67 ms │              9.17 / 9.53 ±0.28 / 9.97 ms │     no change │
│ QQuery 35 │        68.48 / 69.10 ±0.68 / 70.43 ms │           69.20 / 70.28 ±1.55 / 73.36 ms │     no change │
│ QQuery 36 │           5.56 / 5.74 ±0.28 / 6.29 ms │              5.59 / 5.77 ±0.24 / 6.23 ms │     no change │
│ QQuery 37 │           7.20 / 7.36 ±0.29 / 7.94 ms │              7.10 / 7.21 ±0.08 / 7.29 ms │     no change │
│ QQuery 38 │        63.31 / 64.10 ±1.08 / 66.18 ms │           63.27 / 64.66 ±1.72 / 68.05 ms │     no change │
│ QQuery 39 │        70.70 / 72.04 ±1.76 / 75.50 ms │           70.55 / 71.47 ±0.61 / 72.15 ms │     no change │
│ QQuery 40 │        24.43 / 24.84 ±0.47 / 25.70 ms │           22.08 / 23.25 ±1.45 / 25.95 ms │ +1.07x faster │
│ QQuery 41 │        14.65 / 14.77 ±0.14 / 15.03 ms │           13.55 / 13.67 ±0.14 / 13.93 ms │ +1.08x faster │
│ QQuery 42 │        24.51 / 25.70 ±0.96 / 27.46 ms │           23.57 / 24.66 ±0.59 / 25.20 ms │     no change │
│ QQuery 43 │           5.51 / 5.56 ±0.04 / 5.61 ms │              5.57 / 5.63 ±0.05 / 5.71 ms │     no change │
│ QQuery 44 │           9.19 / 9.41 ±0.19 / 9.68 ms │              9.37 / 9.52 ±0.10 / 9.64 ms │     no change │
│ QQuery 45 │        37.67 / 38.41 ±0.64 / 39.60 ms │           37.91 / 39.17 ±1.69 / 42.43 ms │     no change │
│ QQuery 46 │        11.45 / 12.70 ±2.16 / 17.02 ms │           11.79 / 12.00 ±0.14 / 12.15 ms │ +1.06x faster │
│ QQuery 47 │     226.18 / 229.67 ±2.32 / 232.51 ms │        198.29 / 208.09 ±5.83 / 214.70 ms │ +1.10x faster │
│ QQuery 48 │        88.66 / 89.89 ±1.05 / 91.37 ms │           89.41 / 89.77 ±0.21 / 90.05 ms │     no change │
│ QQuery 49 │        68.47 / 69.37 ±0.71 / 70.26 ms │           69.09 / 70.06 ±0.91 / 71.73 ms │     no change │
│ QQuery 50 │        56.41 / 57.83 ±1.09 / 59.46 ms │           55.95 / 56.37 ±0.31 / 56.74 ms │     no change │
│ QQuery 51 │        92.50 / 94.32 ±1.36 / 95.96 ms │           93.58 / 94.73 ±0.84 / 95.86 ms │     no change │
│ QQuery 52 │        24.87 / 24.99 ±0.13 / 25.19 ms │           23.27 / 23.45 ±0.27 / 23.99 ms │ +1.07x faster │
│ QQuery 53 │        29.71 / 31.27 ±2.41 / 36.07 ms │           28.26 / 30.20 ±3.58 / 37.35 ms │     no change │
│ QQuery 54 │        54.69 / 55.67 ±0.94 / 57.33 ms │           51.76 / 52.59 ±0.61 / 53.30 ms │ +1.06x faster │
│ QQuery 55 │        23.09 / 23.79 ±0.61 / 24.64 ms │           22.68 / 22.97 ±0.15 / 23.08 ms │     no change │
│ QQuery 56 │        37.81 / 38.02 ±0.28 / 38.55 ms │           37.99 / 38.60 ±0.58 / 39.69 ms │     no change │
│ QQuery 57 │     148.99 / 151.62 ±2.57 / 155.10 ms │        150.68 / 155.25 ±2.76 / 158.85 ms │     no change │
│ QQuery 58 │      99.97 / 101.56 ±1.32 / 103.77 ms │        104.63 / 106.54 ±1.52 / 109.08 ms │     no change │
│ QQuery 59 │     104.91 / 105.76 ±1.59 / 108.94 ms │        105.62 / 107.81 ±1.90 / 110.71 ms │     no change │
│ QQuery 60 │        41.64 / 41.99 ±0.43 / 42.80 ms │           42.09 / 42.54 ±0.40 / 43.12 ms │     no change │
│ QQuery 61 │          9.03 / 9.83 ±1.26 / 12.31 ms │              8.94 / 9.13 ±0.27 / 9.66 ms │ +1.08x faster │
│ QQuery 62 │        37.04 / 37.49 ±0.39 / 38.20 ms │           37.09 / 37.72 ±0.33 / 38.06 ms │     no change │
│ QQuery 63 │        28.26 / 28.48 ±0.19 / 28.80 ms │           28.67 / 30.05 ±2.47 / 34.98 ms │  1.06x slower │
│ QQuery 64 │     345.50 / 351.18 ±5.97 / 360.75 ms │        350.59 / 354.36 ±2.83 / 358.17 ms │     no change │
│ QQuery 65 │        74.26 / 75.29 ±0.90 / 76.81 ms │           75.15 / 80.17 ±4.16 / 85.18 ms │  1.06x slower │
│ QQuery 66 │        70.62 / 70.88 ±0.30 / 71.38 ms │           75.98 / 76.57 ±0.60 / 77.63 ms │  1.08x slower │
│ QQuery 67 │     215.99 / 220.54 ±2.66 / 223.41 ms │        246.45 / 249.23 ±2.51 / 252.22 ms │  1.13x slower │
│ QQuery 68 │        10.89 / 11.08 ±0.30 / 11.68 ms │           12.29 / 12.42 ±0.09 / 12.54 ms │  1.12x slower │
│ QQuery 69 │        52.71 / 53.32 ±0.99 / 55.30 ms │           55.72 / 58.36 ±3.55 / 65.31 ms │  1.09x slower │
│ QQuery 70 │       95.65 / 98.77 ±2.19 / 101.93 ms │         98.15 / 100.08 ±1.07 / 101.16 ms │     no change │
│ QQuery 71 │        33.42 / 34.46 ±0.65 / 35.27 ms │           34.23 / 34.80 ±0.33 / 35.13 ms │     no change │
│ QQuery 72 │ 1453.17 / 1520.36 ±57.31 / 1624.93 ms │    1580.21 / 1630.52 ±29.90 / 1656.50 ms │  1.07x slower │
│ QQuery 73 │           8.63 / 9.08 ±0.44 / 9.73 ms │             8.83 / 9.81 ±1.49 / 12.75 ms │  1.08x slower │
│ QQuery 74 │     234.57 / 238.68 ±2.61 / 241.49 ms │        238.39 / 242.03 ±2.69 / 245.56 ms │     no change │
│ QQuery 75 │     136.87 / 137.60 ±0.67 / 138.79 ms │        137.05 / 137.63 ±0.76 / 139.00 ms │     no change │
│ QQuery 76 │        33.71 / 34.27 ±0.44 / 34.81 ms │           33.53 / 35.42 ±3.04 / 41.46 ms │     no change │
│ QQuery 77 │        58.01 / 59.12 ±0.81 / 60.47 ms │           58.35 / 61.22 ±1.93 / 63.07 ms │     no change │
│ QQuery 78 │     154.32 / 159.78 ±4.32 / 164.93 ms │        175.89 / 181.30 ±7.20 / 195.36 ms │  1.13x slower │
│ QQuery 79 │        64.93 / 66.98 ±2.79 / 72.50 ms │           71.89 / 72.26 ±0.34 / 72.84 ms │  1.08x slower │
│ QQuery 80 │       91.67 / 95.10 ±2.81 / 100.22 ms │        100.82 / 103.24 ±2.96 / 108.93 ms │  1.09x slower │
│ QQuery 81 │        26.98 / 27.39 ±0.34 / 27.92 ms │           30.54 / 30.88 ±0.21 / 31.10 ms │  1.13x slower │
│ QQuery 82 │        16.50 / 16.57 ±0.06 / 16.68 ms │           18.33 / 18.62 ±0.17 / 18.78 ms │  1.12x slower │
│ QQuery 83 │        30.66 / 31.09 ±0.33 / 31.57 ms │           31.85 / 33.62 ±1.23 / 35.02 ms │  1.08x slower │
│ QQuery 84 │        29.76 / 30.67 ±1.24 / 33.14 ms │           29.85 / 30.88 ±1.49 / 33.83 ms │     no change │
│ QQuery 85 │       92.91 / 95.61 ±3.30 / 101.76 ms │          93.44 / 96.26 ±3.59 / 103.36 ms │     no change │
│ QQuery 86 │        23.44 / 23.91 ±0.53 / 24.66 ms │           24.06 / 24.38 ±0.33 / 24.98 ms │     no change │
│ QQuery 87 │        63.20 / 63.47 ±0.23 / 63.72 ms │           63.66 / 64.17 ±0.41 / 64.71 ms │     no change │
│ QQuery 88 │        57.32 / 58.15 ±0.70 / 59.38 ms │           59.97 / 60.76 ±0.51 / 61.56 ms │     no change │
│ QQuery 89 │        34.61 / 35.02 ±0.46 / 35.87 ms │           36.51 / 38.17 ±2.59 / 43.34 ms │  1.09x slower │
│ QQuery 90 │        15.47 / 16.47 ±1.73 / 19.92 ms │           15.48 / 15.83 ±0.39 / 16.33 ms │     no change │
│ QQuery 91 │        37.47 / 38.04 ±0.41 / 38.54 ms │           37.89 / 38.09 ±0.16 / 38.31 ms │     no change │
│ QQuery 92 │        27.59 / 28.18 ±0.57 / 29.13 ms │           27.78 / 28.30 ±0.27 / 28.58 ms │     no change │
│ QQuery 93 │        47.66 / 48.53 ±0.98 / 50.44 ms │           47.56 / 49.99 ±2.21 / 54.02 ms │     no change │
│ QQuery 94 │        36.57 / 36.99 ±0.53 / 38.02 ms │           36.46 / 36.93 ±0.61 / 38.12 ms │     no change │
│ QQuery 95 │        77.83 / 79.04 ±1.28 / 81.51 ms │           77.71 / 79.20 ±0.91 / 80.50 ms │     no change │
│ QQuery 96 │        22.83 / 23.31 ±0.38 / 23.74 ms │           23.02 / 23.35 ±0.30 / 23.89 ms │     no change │
│ QQuery 97 │        49.78 / 50.77 ±0.62 / 51.57 ms │           49.63 / 50.44 ±0.88 / 51.77 ms │     no change │
│ QQuery 98 │        36.77 / 37.81 ±0.85 / 39.19 ms │           37.22 / 37.55 ±0.31 / 38.07 ms │     no change │
│ QQuery 99 │        55.45 / 56.57 ±1.87 / 60.30 ms │           55.54 / 57.25 ±2.54 / 62.28 ms │     no change │
└───────────┴───────────────────────────────────────┴──────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                       ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                       │ 9318.89ms │
│ Total Time (rich-T-kid_dictionary-aggregate-emission)   │ 9753.93ms │
│ Average Time (HEAD)                                     │   94.13ms │
│ Average Time (rich-T-kid_dictionary-aggregate-emission) │   98.52ms │
│ Queries Faster                                          │        12 │
│ Queries Slower                                          │        25 │
│ Queries with No Change                                  │        62 │
│ Queries with Failure                                    │         0 │
└─────────────────────────────────────────────────────────┴───────────┘

Memory Pool Peaks

Peak MemoryPool reservation per query — what DataFusion's accounting believes it reserved. Recorded only for benchmarks that write a results JSON (the dfbench suites, and the suites bench.sh runs through the Criterion SQL harness since #25644), and only under a memory limit: DATAFUSION_RUNTIME_MEMORY_LIMIT, or one the suite sets itself.

Base: 32bd81fb226ccf1ae5df67ff340d77baa3e73668 | Changed: 939ae6c37e6326f2737600e4d60ef2d3673f119a

tpcds — tpcds_sf1

Query	Base	Changed	Change
Query 1	135.4 KiB	135.4 KiB	+0.0%
Query 2	91.8 MiB	82.9 MiB	-9.7%
Query 3	1.6 MiB	1.9 MiB	+14.5%
Query 4	79.7 MiB	79.7 MiB	-0.0%
Query 5	12.5 MiB	13.0 MiB	+4.2%
Query 6	4.4 MiB	4.7 MiB	+6.6%
Query 7	3.7 MiB	3.7 MiB	+0.9%
Query 8	2.1 MiB	1.7 MiB	-20.6%
Query 9	32.1 KiB	32.1 KiB	+0.0%
Query 10	2.0 MiB	2.0 MiB	-0.5%
Query 11	62.0 MiB	62.0 MiB	-0.1%
Query 12	114.4 MiB	114.4 MiB	+0.0%
Query 13	13.0 MiB	13.7 MiB	+5.1%
Query 14	171.0 MiB	171.6 MiB	+0.3%
Query 15	4.1 MiB	4.2 MiB	+3.9%
Query 16	118.6 KiB	118.6 KiB	+0.0%
Query 17	41.0 MiB	40.6 MiB	-1.1%
Query 18	21.2 MiB	21.2 MiB	-0.0%
Query 19	1.9 MiB	2.0 MiB	+5.7%
Query 20	114.6 MiB	114.6 MiB	-0.0%
Query 21	4.4 MiB	2.9 MiB	-33.3%
Query 22	61.8 MiB	50.5 MiB	-18.2%
Query 23	211.8 MiB	211.7 MiB	-0.1%
Query 24	38.7 MiB	38.7 MiB	-0.1%
Query 25	99.9 MiB	99.9 MiB	-0.0%
Query 26	3.6 MiB	3.5 MiB	-4.2%
Query 27	1.6 MiB	1.6 MiB	-0.3%
Query 28	1.3 MiB	1.3 MiB	+0.0%
Query 29	104.7 MiB	104.7 MiB	-0.0%
Query 30	3.7 MiB	3.7 MiB	-0.8%
Query 31	13.4 MiB	13.4 MiB	-0.0%
Query 32	2.7 MiB	2.5 MiB	-8.1%
Query 33	5.3 MiB	5.3 MiB	-0.0%
Query 34	6.0 MiB	6.0 MiB	+0.0%
Query 35	17.5 MiB	17.6 MiB	+0.3%
Query 36	10.3 MiB	10.3 MiB	+0.2%
Query 37	3.3 MiB	3.3 MiB	-0.1%
Query 38	11.3 MiB	10.9 MiB	-3.8%
Query 39	42.4 MiB	42.4 MiB	+0.0%
Query 40	12.2 MiB	12.2 MiB	-0.0%
Query 41	2.1 MiB	2.1 MiB	-0.2%
Query 42	1.3 MiB	1.5 MiB	+17.2%
Query 43	257.8 KiB	254.7 KiB	-1.2%
Query 44	233.2 KiB	233.2 KiB	+0.0%
Query 45	4.5 MiB	4.5 MiB	-0.1%
Query 46	6.2 MiB	6.2 MiB	-0.0%
Query 47	132.6 MiB	133.1 MiB	+0.4%
Query 48	11.8 MiB	11.8 MiB	+0.0%
Query 49	62.0 MiB	61.1 MiB	-1.3%
Query 50	16.9 MiB	18.0 MiB	+6.5%
Query 51	177.7 MiB	177.9 MiB	+0.1%
Query 52	1.8 MiB	1.5 MiB	-13.1%
Query 53	20.6 MiB	20.6 MiB	+0.0%
Query 54	2.2 MiB	2.1 MiB	-4.2%
Query 55	1.7 MiB	1.5 MiB	-12.3%
Query 56	3.1 MiB	3.0 MiB	-1.3%
Query 57	120.8 MiB	127.7 MiB	+5.8%
Query 58	4.2 MiB	4.2 MiB	-0.1%
Query 59	11.7 MiB	11.7 MiB	+0.0%
Query 60	7.2 MiB	7.2 MiB	-0.1%
Query 61	2.4 KiB	2.5 KiB	+3.3%
Query 62	3.7 MiB	3.7 MiB	+2.0%
Query 63	30.6 MiB	30.4 MiB	-0.5%
Query 64	83.7 MiB	82.8 MiB	-1.1%
Query 65	29.7 MiB	29.7 MiB	+0.0%
Query 66	14.9 MiB	15.0 MiB	+0.1%
Query 67	138.1 MiB	142.4 MiB	+3.1%
Query 68	6.2 MiB	6.2 MiB	-0.0%
Query 69	3.8 MiB	3.8 MiB	-0.3%
Query 70	44.0 MiB	44.0 MiB	-0.0%
Query 71	80.4 MiB	40.6 MiB	-49.5%
Query 72	97.2 MiB	97.2 MiB	-0.0%
Query 73	6.0 MiB	6.0 MiB	+0.0%
Query 74	34.6 MiB	34.6 MiB	-0.0%
Query 75	19.6 MiB	19.7 MiB	+0.3%
Query 76	7.9 MiB	7.9 MiB	-0.1%
Query 77	4.1 MiB	4.1 MiB	-0.0%
Query 78	111.5 MiB	111.1 MiB	-0.3%
Query 79	8.7 MiB	9.3 MiB	+7.5%
Query 80	41.7 MiB	42.0 MiB	+0.7%
Query 81	4.0 MiB	3.9 MiB	-2.3%
Query 82	4.9 MiB	4.9 MiB	-0.1%
Query 83	8.6 MiB	9.7 MiB	+13.5%
Query 84	11.9 MiB	11.9 MiB	+0.0%
Query 85	17.0 MiB	17.0 MiB	-0.0%
Query 86	82.1 MiB	82.1 MiB	+0.0%
Query 87	12.9 MiB	12.7 MiB	-1.9%
Query 88	1.6 MiB	1.6 MiB	-3.2%
Query 89	112.0 MiB	112.2 MiB	+0.2%
Query 90	348.6 KiB	348.6 KiB	+0.0%
Query 91	21.0 MiB	31.1 MiB	+47.7%
Query 92	2.7 MiB	2.7 MiB	+0.0%
Query 93	15.4 MiB	15.1 MiB	-1.6%
Query 94	3.7 MiB	3.7 MiB	+0.0%
Query 95	17.8 MiB	18.8 MiB	+6.0%
Query 96	224.7 KiB	224.7 KiB	+0.0%
Query 97	33.1 MiB	33.1 MiB	+0.0%
Query 98	114.7 MiB	114.7 MiB	+0.0%
Query 99	1.4 MiB	1.7 MiB	+19.0%

Pool accounting vs. process RSS

Max pool peak is the largest reservation any single query in the run reached; peak RSS covers the whole invocation, including data loading and allocator retention, and the two high-water marks need not coincide in time. The gap is therefore an upper bound on what the pool did not account for, not a measurement of it.

Benchmark	Side	Max pool peak	Peak RSS	Gap	RSS / pool
tpcds	base (`32bd81fb226ccf1ae5df67ff340d77baa3e73668`)	211.8 MiB	2.0 GiB	1.7 GiB	9.5×
tpcds	changed (`939ae6c37e6326f2737600e4d60ef2d3673f119a`)	211.7 MiB	1.8 GiB	1.6 GiB	8.7×

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	50.0s
Peak memory	2.0 GiB
Avg memory	1.3 GiB
CPU user	183.1s
CPU sys	5.5s
Peak spill	0 B

tpcds — branch

Metric	Value
Wall time	50.0s
Peak memory	1.8 GiB
Avg memory	1.3 GiB
CPU user	192.7s
CPU sys	5.8s
Peak spill	0 B

File an issue against this benchmark runner

Rich-T-kid · 2026-11-03T15:10:38Z

run benchmark dictionary_group_values

baseline:
ref: "32bd81fb226ccf1ae5df67ff340d77baa3e73668"
changed:
ref: "4653896c9a2174492c86651ae5cf28172795f28b"

Rich-T-kid · 2026-11-03T15:10:39Z

run benchmark dictionary_group_values

baseline:
ref: "32bd81fb226ccf1ae5df67ff340d77baa3e73668"
changed:
ref: "4653896c9a2174492c86651ae5cf28172795f28b"

Rich-T-kid · 2026-11-03T15:10:40Z

run benchmark dictionary_group_values

baseline:
ref: "32bd81fb226ccf1ae5df67ff340d77baa3e73668"
changed:
ref: "4653896c9a2174492c86651ae5cf28172795f28b"

Rich-T-kid · 2026-11-03T15:13:48Z

show benchmark queue

adriangbot · 2026-11-03T15:13:51Z

Hi @Rich-T-kid, you asked to view the benchmark queue (#27051 (comment)).

Comment	Repo	PR	User	Benchmarks	Status
#6607440604	apache/datafusion	#27051	Rich-T-kid	["dictionary_group_values"]	running
#6607440040	apache/datafusion	#27051	Rich-T-kid	["dictionary_group_values"]	running
#6607439926	apache/datafusion	#27051	Rich-T-kid	["dictionary_group_values"]	running

File an issue against this benchmark runner

adriangbot · 2026-11-03T15:13:53Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6607440604-4306-xjd5n 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing 4653896 (4653896) to 32bd81f diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "32bd81fb226ccf1ae5df67ff340d77baa3e73668"
changed:
  ref: "4653896c9a2174492c86651ae5cf28172795f28b"

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-11-03T15:14:08Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6607439926-4308-x8ztx 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing 4653896 (4653896) to 32bd81f diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "32bd81fb226ccf1ae5df67ff340d77baa3e73668"
changed:
  ref: "4653896c9a2174492c86651ae5cf28172795f28b"

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-11-03T15:14:40Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6607440040-4307-2j2qh 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing 4653896 (4653896) to 32bd81f diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "32bd81fb226ccf1ae5df67ff340d77baa3e73668"
changed:
  ref: "4653896c9a2174492c86651ae5cf28172795f28b"

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-11-03T15:39:18Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing 4653896 (4653896) to 32bd81f diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "32bd81fb226ccf1ae5df67ff340d77baa3e73668"
changed:
  ref: "4653896c9a2174492c86651ae5cf28172795f28b"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

group                                                                             HEAD                                    rich-T-kid_dictionary-aggregate-emission
-----                                                                             ----                                    ----------------------------------------
dict_intern_emit/intern_emit/size_65536_card_1000_null_0.00                       1.00    478.6±3.82µs 130.6 MElem/sec    1.00    478.6±2.62µs 130.6 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_20_null_0.00                         1.00    414.6±3.00µs 150.8 MElem/sec    1.00    413.9±3.25µs 151.0 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_300_null_0.00                        1.00    442.2±3.69µs 141.4 MElem/sec    1.00    440.1±2.85µs 142.0 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_65536_null_0.00                      1.00      6.2±0.04ms 10.1 MElem/sec     1.04      6.5±0.01ms  9.7 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_75_null_0.00                         1.00    432.7±2.91µs 144.5 MElem/sec    1.00    431.4±3.61µs 144.9 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_1000_null_0.00                        1.00    112.7±5.14µs 69.3 MElem/sec     1.00    113.0±0.89µs 69.2 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_20_null_0.00                          1.01     55.9±0.45µs 139.8 MElem/sec    1.00     55.3±0.71µs 141.3 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_300_null_0.00                         1.00     72.5±1.61µs 107.8 MElem/sec    1.09    79.0±64.49µs 98.9 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_75_null_0.00                          1.02     60.9±0.69µs 128.4 MElem/sec    1.00     59.9±0.61µs 130.3 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_8192_null_0.00                        1.00    524.8±5.49µs 14.9 MElem/sec     1.04    547.2±2.24µs 14.3 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_1000_null_0.10     1.01      2.3±0.01ms 107.2 MElem/sec    1.00      2.3±0.01ms 108.1 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_20_null_0.10       1.00      2.2±0.00ms 116.2 MElem/sec    1.00      2.1±0.00ms 116.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_300_null_0.10      1.00      2.2±0.01ms 111.5 MElem/sec    1.00      2.2±0.01ms 111.7 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_65536_null_0.10    1.10     19.0±0.08ms 13.1 MElem/sec     1.00     17.4±0.10ms 14.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_75_null_0.10       1.00      2.2±0.00ms 112.0 MElem/sec    1.00      2.2±0.00ms 112.3 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_1000_null_0.10      1.03    395.6±2.35µs 79.0 MElem/sec     1.00    384.1±1.91µs 81.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_20_null_0.10        1.00    248.1±3.21µs 126.0 MElem/sec    1.02    251.9±1.61µs 124.1 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_300_null_0.10       1.00    293.9±2.85µs 106.3 MElem/sec    1.00    293.3±1.98µs 106.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_75_null_0.10        1.00    263.4±2.19µs 118.6 MElem/sec    1.02    267.6±1.61µs 116.8 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_8192_null_0.10      1.08   1310.5±5.91µs 23.8 MElem/sec     1.00   1218.4±5.97µs 25.6 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_1000_null_0.10       1.04   1353.5±2.83µs 46.2 MElem/sec     1.00   1301.3±5.16µs 48.0 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_20_null_0.10         1.03   1221.7±3.83µs 51.2 MElem/sec     1.00   1185.8±2.28µs 52.7 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_300_null_0.10        1.04   1298.3±3.90µs 48.1 MElem/sec     1.00   1250.4±2.62µs 50.0 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_65536_null_0.10      1.14      7.8±0.02ms  8.0 MElem/sec     1.00      6.8±0.04ms  9.1 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_75_null_0.10         1.00   1246.3±3.05µs 50.2 MElem/sec     1.00   1240.1±2.72µs 50.4 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_100000                    1.00     18.9±0.07ms 13.2 MElem/sec     1.06     20.0±0.08ms 12.5 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_32768                     1.00      7.9±0.01ms 31.8 MElem/sec     1.04      8.2±0.01ms 30.6 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_500000                    1.00     64.9±0.89ms  3.9 MElem/sec     1.08     69.9±0.95ms  3.6 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_65536                     1.00     14.1±0.04ms 17.7 MElem/sec     1.05     14.7±0.03ms 17.0 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_8192                      1.00      2.8±0.02ms 88.8 MElem/sec     1.00      2.8±0.02ms 88.9 MElem/sec
dict_take_n/take_n/size_65536_card_1000_null_0.10                                 1.00      2.5±0.01ms 102.0 MElem/sec    1.00      2.5±0.01ms 101.5 MElem/sec
dict_take_n/take_n/size_65536_card_20_null_0.10                                   1.00      2.1±0.01ms 120.8 MElem/sec    1.01      2.1±0.01ms 119.2 MElem/sec
dict_take_n/take_n/size_65536_card_300_null_0.10                                  1.00      2.2±0.00ms 112.1 MElem/sec    1.01      2.2±0.00ms 111.2 MElem/sec
dict_take_n/take_n/size_65536_card_65536_null_0.10                                1.00     26.5±0.06ms  9.4 MElem/sec     1.02     27.0±0.06ms  9.3 MElem/sec
dict_take_n/take_n/size_65536_card_75_null_0.10                                   1.00      2.2±0.01ms 115.9 MElem/sec    1.02      2.2±0.01ms 113.7 MElem/sec
group_values_emit/dictionary/size_500000_null_0.0                                                                         1.00      3.2±0.08ms 150.0 MElem/sec
group_values_emit/dictionary/size_500000_null_0.1                                                                         1.00      2.6±0.13ms 184.5 MElem/sec
group_values_emit/dictionary/size_65536_null_0.0                                                                          1.00      3.4±0.26µs 18.0 GElem/sec
group_values_emit/dictionary/size_65536_null_0.1                                                                          1.00      5.1±0.37µs 11.9 GElem/sec
group_values_emit/dictionary/size_8192_null_0.0                                                                           1.00     4.8±21.93µs 1622.8 MElem/sec
group_values_emit/dictionary/size_8192_null_0.1                                                                           1.00     4.8±16.76µs 1644.2 MElem/sec
group_values_emit/utf8_view/size_500000_null_0.0                                                                          1.00  1976.1±44.12µs 241.3 MElem/sec
group_values_emit/utf8_view/size_500000_null_0.1                                                                          1.00      7.5±0.43µs 61.9 GElem/sec
group_values_emit/utf8_view/size_65536_null_0.0                                                                           1.00  1231.8±106.95ns 49.6 GElem/sec
group_values_emit/utf8_view/size_65536_null_0.1                                                                           1.00      2.1±0.12µs 29.2 GElem/sec
group_values_emit/utf8_view/size_8192_null_0.0                                                                            1.00   597.5±55.81ns 12.8 GElem/sec
group_values_emit/utf8_view/size_8192_null_0.1                                                                            1.00  872.2±100.96ns  8.7 GElem/sec

Resource Usage

dictionary_group_values — base (merge-base)

Metric	Value
Wall time	665.1s
Peak memory	901.2 MiB
Avg memory	62.2 MiB
CPU user	407.8s
CPU sys	22.6s
Peak spill	0 B

dictionary_group_values — branch

Metric	Value
Wall time	800.2s
Peak memory	889.3 MiB
Avg memory	78.8 MiB
CPU user	537.4s
CPU sys	42.1s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-11-03T15:39:41Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing 4653896 (4653896) to 32bd81f diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "32bd81fb226ccf1ae5df67ff340d77baa3e73668"
changed:
  ref: "4653896c9a2174492c86651ae5cf28172795f28b"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

group                                                                             HEAD                                    rich-T-kid_dictionary-aggregate-emission
-----                                                                             ----                                    ----------------------------------------
dict_intern_emit/intern_emit/size_65536_card_1000_null_0.00                       1.01    488.9±6.54µs 127.8 MElem/sec    1.00    482.7±6.45µs 129.5 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_20_null_0.00                         1.03    427.9±7.58µs 146.1 MElem/sec    1.00    415.3±5.81µs 150.5 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_300_null_0.00                        1.01    448.1±7.11µs 139.5 MElem/sec    1.00    445.7±6.33µs 140.2 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_65536_null_0.00                      1.00      6.2±0.02ms 10.0 MElem/sec     1.03      6.4±0.02ms  9.8 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_75_null_0.00                         1.02    443.8±5.86µs 140.8 MElem/sec    1.00    435.8±5.09µs 143.4 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_1000_null_0.00                        1.00    113.3±1.29µs 69.0 MElem/sec     1.00    113.8±1.35µs 68.6 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_20_null_0.00                          1.02     57.2±1.29µs 136.6 MElem/sec    1.00     56.2±1.38µs 139.0 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_300_null_0.00                         1.00     73.5±1.17µs 106.3 MElem/sec    1.00     73.4±1.51µs 106.5 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_75_null_0.00                          1.01     61.4±1.11µs 127.2 MElem/sec    1.00     60.7±1.22µs 128.6 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_8192_null_0.00                        1.00    523.8±3.15µs 14.9 MElem/sec     1.05    547.5±3.45µs 14.3 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_1000_null_0.10     1.01      2.3±0.01ms 106.9 MElem/sec    1.00      2.3±0.01ms 107.9 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_20_null_0.10       1.01      2.2±0.01ms 115.3 MElem/sec    1.00      2.1±0.01ms 116.5 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_300_null_0.10      1.01      2.3±0.01ms 110.6 MElem/sec    1.00      2.2±0.01ms 111.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_65536_null_0.10    1.03     20.1±0.39ms 12.4 MElem/sec     1.00     19.4±0.38ms 12.9 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_75_null_0.10       1.01      2.2±0.00ms 111.7 MElem/sec    1.00      2.2±0.01ms 112.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_1000_null_0.10      1.04    398.2±2.93µs 78.5 MElem/sec     1.00    384.7±2.65µs 81.2 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_20_null_0.10        1.00   250.4±20.43µs 124.8 MElem/sec    1.01    252.1±1.52µs 124.0 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_300_null_0.10       1.00    295.7±3.66µs 105.7 MElem/sec    1.00    294.7±6.08µs 106.0 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_75_null_0.10        1.06  295.2±298.83µs 105.8 MElem/sec    1.00  279.1±123.67µs 112.0 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_8192_null_0.10      1.10  1355.8±437.26µs 23.0 MElem/sec    1.00  1230.3±73.36µs 25.4 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_1000_null_0.10       1.02   1357.2±4.03µs 46.0 MElem/sec     1.00  1336.3±29.61µs 46.8 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_20_null_0.10         1.02   1224.3±3.78µs 51.1 MElem/sec     1.00  1194.7±28.28µs 52.3 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_300_null_0.10        1.00   1265.3±5.35µs 49.4 MElem/sec     1.00  1260.0±30.66µs 49.6 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_65536_null_0.10      1.17      7.8±0.04ms  8.0 MElem/sec     1.00      6.7±0.05ms  9.3 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_75_null_0.10         1.06   1283.7±4.38µs 48.7 MElem/sec     1.00  1211.8±28.17µs 51.6 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_100000                    1.00     21.1±0.37ms 11.9 MElem/sec     1.10     23.1±0.47ms 10.8 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_32768                     1.00      7.8±0.02ms 32.1 MElem/sec     1.01      7.9±0.02ms 31.7 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_500000                    1.00     76.2±1.32ms  3.3 MElem/sec     1.08     82.1±1.52ms  3.0 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_65536                     1.00     14.8±0.16ms 16.9 MElem/sec     1.07     15.9±0.22ms 15.8 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_8192                      1.00      2.8±0.02ms 88.5 MElem/sec     1.00      2.8±0.01ms 88.1 MElem/sec
dict_take_n/take_n/size_65536_card_1000_null_0.10                                 1.00      2.5±0.01ms 102.0 MElem/sec    1.01      2.5±0.01ms 101.1 MElem/sec
dict_take_n/take_n/size_65536_card_20_null_0.10                                   1.00      2.1±0.01ms 120.1 MElem/sec    1.00      2.1±0.01ms 119.7 MElem/sec
dict_take_n/take_n/size_65536_card_300_null_0.10                                  1.00      2.2±0.01ms 112.0 MElem/sec    1.01      2.3±0.01ms 111.0 MElem/sec
dict_take_n/take_n/size_65536_card_65536_null_0.10                                1.00     26.6±0.11ms  9.4 MElem/sec     1.03     27.5±0.17ms  9.1 MElem/sec
dict_take_n/take_n/size_65536_card_75_null_0.10                                   1.00      2.2±0.01ms 115.4 MElem/sec    1.01      2.2±0.01ms 114.3 MElem/sec
group_values_emit/dictionary/size_500000_null_0.0                                                                         1.00     11.9±1.62µs 39.1 GElem/sec
group_values_emit/dictionary/size_500000_null_0.1                                                                         1.00      3.0±0.09ms 160.7 MElem/sec
group_values_emit/dictionary/size_65536_null_0.0                                                                          1.00      6.4±0.42µs  9.6 GElem/sec
group_values_emit/dictionary/size_65536_null_0.1                                                                          1.00      8.3±0.61µs  7.3 GElem/sec
group_values_emit/dictionary/size_8192_null_0.0                                                                           1.00     4.1±13.24µs 1914.0 MElem/sec
group_values_emit/dictionary/size_8192_null_0.1                                                                           1.00      2.9±1.57µs  2.7 GElem/sec
group_values_emit/utf8_view/size_500000_null_0.0                                                                          1.00      2.5±0.09ms 187.7 MElem/sec
group_values_emit/utf8_view/size_500000_null_0.1                                                                          1.00      8.7±0.49µs 53.5 GElem/sec
group_values_emit/utf8_view/size_65536_null_0.0                                                                           1.00  1715.4±155.25ns 35.6 GElem/sec
group_values_emit/utf8_view/size_65536_null_0.1                                                                           1.00      2.7±0.20µs 23.0 GElem/sec
group_values_emit/utf8_view/size_8192_null_0.0                                                                            1.00   652.2±91.34ns 11.7 GElem/sec
group_values_emit/utf8_view/size_8192_null_0.1                                                                            1.00  1060.0±151.56ns  7.2 GElem/sec

Resource Usage

dictionary_group_values — base (merge-base)

Metric	Value
Wall time	720.2s
Peak memory	901.2 MiB
Avg memory	59.1 MiB
CPU user	412.9s
CPU sys	24.1s
Peak spill	0 B

dictionary_group_values — branch

Metric	Value
Wall time	815.2s
Peak memory	719.4 MiB
Avg memory	83.3 MiB
CPU user	553.8s
CPU sys	40.6s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-11-03T15:39:52Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing 4653896 (4653896) to 32bd81f diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "32bd81fb226ccf1ae5df67ff340d77baa3e73668"
changed:
  ref: "4653896c9a2174492c86651ae5cf28172795f28b"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

group                                                                             HEAD                                    rich-T-kid_dictionary-aggregate-emission
-----                                                                             ----                                    ----------------------------------------
dict_intern_emit/intern_emit/size_65536_card_1000_null_0.00                       1.00    488.6±7.00µs 127.9 MElem/sec    1.00    488.5±6.68µs 127.9 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_20_null_0.00                         1.00    422.3±4.42µs 148.0 MElem/sec    1.00    420.6±7.17µs 148.6 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_300_null_0.00                        1.00    448.9±5.33µs 139.2 MElem/sec    1.01    452.4±6.30µs 138.2 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_65536_null_0.00                      1.00      6.1±0.04ms 10.3 MElem/sec     1.09      6.6±0.06ms  9.4 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_75_null_0.00                         1.01    441.8±6.30µs 141.5 MElem/sec    1.00    439.0±5.34µs 142.4 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_1000_null_0.00                        1.00    114.0±3.75µs 68.5 MElem/sec     1.00    114.2±1.52µs 68.4 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_20_null_0.00                          1.00     56.4±1.01µs 138.6 MElem/sec    1.01     57.0±1.30µs 137.1 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_300_null_0.00                         1.00     72.7±0.88µs 107.5 MElem/sec    1.01     73.6±1.66µs 106.1 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_75_null_0.00                          1.02     62.0±1.24µs 126.0 MElem/sec    1.00     61.0±1.41µs 128.1 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_8192_null_0.00                        1.00    527.8±2.73µs 14.8 MElem/sec     1.06   557.2±63.62µs 14.0 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_1000_null_0.10     1.01      2.3±0.01ms 106.7 MElem/sec    1.00      2.3±0.02ms 107.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_20_null_0.10       1.01      2.2±0.00ms 115.4 MElem/sec    1.00      2.2±0.01ms 116.0 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_300_null_0.10      1.01      2.3±0.01ms 110.4 MElem/sec    1.00      2.2±0.01ms 111.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_65536_null_0.10    1.00     20.9±0.79ms 12.0 MElem/sec     1.03     21.6±0.58ms 11.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_75_null_0.10       1.00      2.2±0.01ms 112.2 MElem/sec    1.00      2.2±0.01ms 111.9 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_1000_null_0.10      1.03    398.7±4.23µs 78.4 MElem/sec     1.00    386.7±4.00µs 80.8 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_20_null_0.10        1.00    249.5±2.49µs 125.3 MElem/sec    1.02    253.2±1.79µs 123.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_300_null_0.10       1.01    295.9±3.57µs 105.6 MElem/sec    1.00    294.0±1.89µs 106.3 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_75_null_0.10        1.09  293.0±266.43µs 106.6 MElem/sec    1.00    267.7±1.84µs 116.8 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_8192_null_0.10      1.10  1345.9±247.22µs 23.2 MElem/sec    1.00   1219.1±9.07µs 25.6 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_1000_null_0.10       1.04   1358.6±4.67µs 46.0 MElem/sec     1.00   1302.1±3.38µs 48.0 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_20_null_0.10         1.06   1262.8±3.90µs 49.5 MElem/sec     1.00   1191.5±3.92µs 52.5 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_300_null_0.10        1.01   1269.0±5.51µs 49.3 MElem/sec     1.00   1253.0±5.64µs 49.9 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_65536_null_0.10      1.14      8.1±0.06ms  7.7 MElem/sec     1.00      7.1±0.07ms  8.8 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_75_null_0.10         1.07   1287.5±4.44µs 48.5 MElem/sec     1.00   1206.8±3.33µs 51.8 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_100000                    1.00     22.0±0.68ms 11.4 MElem/sec     1.11     24.4±1.04ms 10.3 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_32768                     1.00      8.1±0.04ms 31.0 MElem/sec     1.02      8.2±0.08ms 30.4 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_500000                    1.00     81.3±2.95ms  3.1 MElem/sec     1.13     92.0±3.91ms  2.7 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_65536                     1.00     15.8±0.65ms 15.8 MElem/sec     1.01     16.0±0.43ms 15.7 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_8192                      1.00      2.8±0.02ms 87.7 MElem/sec     1.03      2.9±0.02ms 85.6 MElem/sec
dict_take_n/take_n/size_65536_card_1000_null_0.10                                 1.00      2.5±0.01ms 101.1 MElem/sec    1.00      2.5±0.01ms 101.3 MElem/sec
dict_take_n/take_n/size_65536_card_20_null_0.10                                   1.00      2.1±0.01ms 119.8 MElem/sec    1.01      2.1±0.01ms 119.0 MElem/sec
dict_take_n/take_n/size_65536_card_300_null_0.10                                  1.00      2.2±0.01ms 111.9 MElem/sec    1.01      2.3±0.01ms 110.8 MElem/sec
dict_take_n/take_n/size_65536_card_65536_null_0.10                                1.00     27.3±0.42ms  9.2 MElem/sec     1.00     27.4±0.34ms  9.1 MElem/sec
dict_take_n/take_n/size_65536_card_75_null_0.10                                   1.00      2.2±0.01ms 113.8 MElem/sec    1.00      2.2±0.01ms 114.1 MElem/sec
group_values_emit/dictionary/size_500000_null_0.0                                                                         1.00     12.1±1.61µs 38.4 GElem/sec
group_values_emit/dictionary/size_500000_null_0.1                                                                         1.00     17.4±1.60µs 26.7 GElem/sec
group_values_emit/dictionary/size_65536_null_0.0                                                                          1.00      7.3±0.69µs  8.4 GElem/sec
group_values_emit/dictionary/size_65536_null_0.1                                                                          1.00      8.1±0.81µs  7.6 GElem/sec
group_values_emit/dictionary/size_8192_null_0.0                                                                           1.00   55.8±521.94µs 140.0 MElem/sec
group_values_emit/dictionary/size_8192_null_0.1                                                                           1.00      3.7±5.82µs  2.1 GElem/sec
group_values_emit/utf8_view/size_500000_null_0.0                                                                          1.00      5.0±0.46µs 94.0 GElem/sec
group_values_emit/utf8_view/size_500000_null_0.1                                                                          1.00      9.2±0.49µs 50.8 GElem/sec
group_values_emit/utf8_view/size_65536_null_0.0                                                                           1.00  1912.6±200.73ns 31.9 GElem/sec
group_values_emit/utf8_view/size_65536_null_0.1                                                                           1.00      2.4±0.16µs 25.2 GElem/sec
group_values_emit/utf8_view/size_8192_null_0.0                                                                            1.00  736.1±169.29ns 10.4 GElem/sec
group_values_emit/utf8_view/size_8192_null_0.1                                                                            1.00  1258.1±174.12ns  6.1 GElem/sec

Resource Usage

dictionary_group_values — base (merge-base)

Metric	Value
Wall time	700.2s
Peak memory	900.9 MiB
Avg memory	59.3 MiB
CPU user	414.9s
CPU sys	23.3s
Peak spill	0 B

dictionary_group_values — branch

Metric	Value
Wall time	830.2s
Peak memory	812.7 MiB
Avg memory	99.3 MiB
CPU user	566.7s
CPU sys	33.1s
Peak spill	0 B

File an issue against this benchmark runner

Rich-T-kid · 2026-11-03T15:57:20Z

run benchmark dictionary_group_values

baseline:
ref: "522287c20d3c09c1b2bd2a8a39bc3a164257614e"
changed:
ref: "b2825948e531b1afe34769f29909359216175b50"

Rich-T-kid · 2026-11-03T15:57:21Z

run benchmark dictionary_group_values

baseline:
ref: "522287c20d3c09c1b2bd2a8a39bc3a164257614e"
changed:
ref: "b2825948e531b1afe34769f29909359216175b50"

Rich-T-kid · 2026-11-03T15:57:21Z

run benchmark dictionary_group_values

baseline:
ref: "522287c20d3c09c1b2bd2a8a39bc3a164257614e"
changed:
ref: "b2825948e531b1afe34769f29909359216175b50"

adriangbot · 2026-11-03T16:01:10Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6608686523-4313-8xjhm 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing b282594 (b282594) to 522287c diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "522287c20d3c09c1b2bd2a8a39bc3a164257614e"
changed:
  ref: "b2825948e531b1afe34769f29909359216175b50"

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-11-03T16:01:11Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6608686227-4312-8lxwc 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing b282594 (b282594) to 522287c diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "522287c20d3c09c1b2bd2a8a39bc3a164257614e"
changed:
  ref: "b2825948e531b1afe34769f29909359216175b50"

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-11-03T16:01:11Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6608685951-4314-clwbz 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing b282594 (b282594) to 522287c diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "522287c20d3c09c1b2bd2a8a39bc3a164257614e"
changed:
  ref: "b2825948e531b1afe34769f29909359216175b50"

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-11-03T16:20:54Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing b282594 (b282594) to 522287c diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "522287c20d3c09c1b2bd2a8a39bc3a164257614e"
changed:
  ref: "b2825948e531b1afe34769f29909359216175b50"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

group                                                                             HEAD                                     rich-T-kid_dictionary-aggregate-emission
-----                                                                             ----                                     ----------------------------------------
dict_intern_emit/intern_emit/size_65536_card_1000_null_0.00                       1.00    482.0±7.82µs 129.7 MElem/sec     1.03    494.6±6.19µs 126.4 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_20_null_0.00                         1.00    418.2±5.78µs 149.4 MElem/sec     1.04    435.8±4.45µs 143.4 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_300_null_0.00                        1.00    443.1±5.69µs 141.0 MElem/sec     1.02    453.8±4.69µs 137.7 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_65536_null_0.00                      1.00      6.3±0.02ms 10.0 MElem/sec      1.03      6.4±0.02ms  9.7 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_75_null_0.00                         1.00    436.3±5.66µs 143.2 MElem/sec     1.03    448.7±4.16µs 139.3 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_1000_null_0.00                        1.00    112.7±1.15µs 69.3 MElem/sec      1.01    114.0±1.14µs 68.6 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_20_null_0.00                          1.00     56.1±1.02µs 139.2 MElem/sec     1.02     57.0±1.08µs 137.1 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_300_null_0.00                         1.00     72.3±0.80µs 108.1 MElem/sec     1.02     73.5±1.09µs 106.2 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_75_null_0.00                          1.00     60.8±0.84µs 128.5 MElem/sec     1.06    64.1±19.40µs 121.8 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_8192_null_0.00                        1.00    523.5±4.13µs 14.9 MElem/sec      1.02    531.7±1.39µs 14.7 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_1000_null_0.10     1.00      2.3±0.00ms 106.9 MElem/sec     1.01      2.4±0.01ms 106.3 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_20_null_0.10       1.00      2.2±0.00ms 115.4 MElem/sec     1.06      2.3±0.00ms 108.8 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_300_null_0.10      1.00      2.2±0.01ms 111.2 MElem/sec     1.03      2.3±0.01ms 108.3 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_65536_null_0.10    1.09     18.5±0.31ms 13.5 MElem/sec      1.00     17.1±0.31ms 14.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_75_null_0.10       1.00      2.2±0.01ms 111.4 MElem/sec     1.02      2.3±0.00ms 109.2 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_1000_null_0.10      1.03    400.0±2.20µs 78.1 MElem/sec      1.00    388.2±2.31µs 80.5 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_20_null_0.10        1.00    263.7±2.14µs 118.5 MElem/sec     1.06    278.6±1.52µs 112.2 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_300_null_0.10       1.00    304.3±2.86µs 102.7 MElem/sec     1.00    304.3±1.49µs 102.7 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_75_null_0.10        1.00    276.7±1.86µs 112.9 MElem/sec     1.03    283.9±1.29µs 110.1 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_8192_null_0.10      1.09   1319.8±5.45µs 23.7 MElem/sec      1.00   1207.9±5.54µs 25.9 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_1000_null_0.10       1.02   1362.3±3.73µs 45.9 MElem/sec      1.00   1337.5±4.04µs 46.7 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_20_null_0.10         1.00   1230.3±4.26µs 50.8 MElem/sec      1.04   1280.5±3.16µs 48.8 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_300_null_0.10        1.03   1269.8±4.71µs 49.2 MElem/sec      1.00   1233.9±2.98µs 50.7 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_65536_null_0.10      1.18      7.6±0.04ms  8.2 MElem/sec      1.00      6.4±0.04ms  9.7 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_75_null_0.10         1.05   1288.1±3.22µs 48.5 MElem/sec      1.00   1232.3±3.06µs 50.7 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_100000                    1.00     19.3±0.24ms 13.0 MElem/sec      1.05     20.2±0.24ms 12.4 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_32768                     1.06      8.3±0.02ms 30.2 MElem/sec      1.00      7.8±0.02ms 32.1 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_500000                    1.00     71.9±1.17ms  3.5 MElem/sec      1.03     74.1±1.30ms  3.4 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_65536                     1.00     14.0±0.07ms 17.9 MElem/sec      1.07     14.9±0.08ms 16.8 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_8192                      1.00      2.8±0.01ms 88.6 MElem/sec      1.00      2.8±0.02ms 88.7 MElem/sec
dict_take_n/take_n/size_65536_card_1000_null_0.10                                 1.00      2.5±0.01ms 100.7 MElem/sec     1.02      2.5±0.00ms 99.1 MElem/sec
dict_take_n/take_n/size_65536_card_20_null_0.10                                   1.00      2.1±0.01ms 116.6 MElem/sec     1.08      2.3±0.01ms 108.3 MElem/sec
dict_take_n/take_n/size_65536_card_300_null_0.10                                  1.00      2.3±0.01ms 109.4 MElem/sec     1.02      2.3±0.01ms 107.4 MElem/sec
dict_take_n/take_n/size_65536_card_65536_null_0.10                                1.02     27.5±0.09ms  9.1 MElem/sec      1.00     26.8±0.09ms  9.3 MElem/sec
dict_take_n/take_n/size_65536_card_75_null_0.10                                   1.00      2.2±0.01ms 112.5 MElem/sec     1.03      2.3±0.00ms 109.0 MElem/sec
group_values_emit/dictionary/size_500000_null_0.0                                 401.59     3.8±0.62ms 126.5 MElem/sec    1.00      9.4±1.05µs 49.6 GElem/sec
group_values_emit/dictionary/size_500000_null_0.1                                 572.49     8.7±0.60ms 54.5 MElem/sec     1.00     15.3±1.72µs 30.5 GElem/sec
group_values_emit/dictionary/size_65536_null_0.0                                  15.57  69.2±112.90µs 903.5 MElem/sec     1.00      4.4±0.45µs 13.7 GElem/sec
group_values_emit/dictionary/size_65536_null_0.1                                  127.11  728.6±10.90µs 85.8 MElem/sec     1.00      5.7±0.58µs 10.6 GElem/sec
group_values_emit/dictionary/size_8192_null_0.0                                   26.77  50.2±304.32µs 155.6 MElem/sec     1.00  1875.9±395.80ns  4.1 GElem/sec
group_values_emit/dictionary/size_8192_null_0.1                                   26.25   99.0±21.18µs 78.9 MElem/sec      1.00      3.8±9.88µs  2.0 GElem/sec
group_values_emit/utf8_view/size_500000_null_0.0                                  535.29     2.2±0.12ms 215.3 MElem/sec    1.00      4.1±0.48µs 112.5 GElem/sec
group_values_emit/utf8_view/size_500000_null_0.1                                  1.02      8.6±0.52µs 54.2 GElem/sec      1.00      8.4±0.77µs 55.3 GElem/sec
group_values_emit/utf8_view/size_65536_null_0.0                                   1.13  1577.1±108.29ns 38.7 GElem/sec     1.00  1397.8±111.65ns 43.7 GElem/sec
group_values_emit/utf8_view/size_65536_null_0.1                                   1.04      2.2±0.14µs 27.9 GElem/sec      1.00      2.1±0.14µs 28.9 GElem/sec
group_values_emit/utf8_view/size_8192_null_0.0                                    1.00   641.2±68.70ns 11.9 GElem/sec      1.07   687.6±82.04ns 11.1 GElem/sec
group_values_emit/utf8_view/size_8192_null_0.1                                    1.00  911.4±125.82ns  8.4 GElem/sec      1.07   977.1±76.98ns  7.8 GElem/sec

Resource Usage

dictionary_group_values — base (merge-base)

Metric	Value
Wall time	540.1s
Peak memory	896.7 MiB
Avg memory	113.5 MiB
CPU user	537.3s
CPU sys	43.2s
Peak spill	0 B

dictionary_group_values — branch

Metric	Value
Wall time	630.1s
Peak memory	889.0 MiB
Avg memory	214.8 MiB
CPU user	575.8s
CPU sys	22.4s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-11-03T16:21:13Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing b282594 (b282594) to 522287c diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "522287c20d3c09c1b2bd2a8a39bc3a164257614e"
changed:
  ref: "b2825948e531b1afe34769f29909359216175b50"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

group                                                                             HEAD                                     rich-T-kid_dictionary-aggregate-emission
-----                                                                             ----                                     ----------------------------------------
dict_intern_emit/intern_emit/size_65536_card_1000_null_0.00                       1.00    487.3±5.87µs 128.3 MElem/sec     1.02    495.5±5.94µs 126.1 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_20_null_0.00                         1.00    422.7±6.13µs 147.9 MElem/sec     1.03    436.1±4.97µs 143.3 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_300_null_0.00                        1.00    445.5±5.12µs 140.3 MElem/sec     1.02    455.5±4.76µs 137.2 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_65536_null_0.00                      1.00      6.3±0.03ms 10.0 MElem/sec      1.03      6.5±0.02ms  9.7 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_75_null_0.00                         1.00    437.4±5.28µs 142.9 MElem/sec     1.03    449.9±4.02µs 138.9 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_1000_null_0.00                        1.00    113.6±1.76µs 68.7 MElem/sec      1.01    114.7±1.04µs 68.1 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_20_null_0.00                          1.00     56.9±1.31µs 137.3 MElem/sec     1.00     57.1±1.21µs 136.9 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_300_null_0.00                         1.00     72.9±1.15µs 107.2 MElem/sec     1.01     73.9±1.21µs 105.8 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_75_null_0.00                          1.00     61.3±1.08µs 127.5 MElem/sec     1.02     62.5±1.00µs 125.0 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_8192_null_0.00                        1.00    527.4±2.97µs 14.8 MElem/sec      1.01    530.7±1.51µs 14.7 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_1000_null_0.10     1.00      2.3±0.01ms 107.3 MElem/sec     1.02      2.4±0.01ms 105.5 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_20_null_0.10       1.00      2.2±0.01ms 115.6 MElem/sec     1.07      2.3±0.00ms 108.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_300_null_0.10      1.00      2.3±0.01ms 110.8 MElem/sec     1.02      2.3±0.01ms 108.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_65536_null_0.10    1.17     21.5±0.65ms 11.6 MElem/sec      1.00     18.4±0.39ms 13.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_75_null_0.10       1.00      2.2±0.01ms 111.5 MElem/sec     1.03      2.3±0.01ms 108.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_1000_null_0.10      1.03    401.0±1.82µs 77.9 MElem/sec      1.00    388.7±1.95µs 80.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_20_null_0.10        1.00    265.9±2.57µs 117.5 MElem/sec     1.05    278.9±2.07µs 112.0 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_300_null_0.10       1.01    306.6±2.89µs 101.9 MElem/sec     1.00    305.1±1.47µs 102.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_75_null_0.10        1.00    277.6±2.27µs 112.6 MElem/sec     1.03    284.9±1.24µs 109.7 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_8192_null_0.10      1.10   1315.0±8.76µs 23.8 MElem/sec      1.00   1200.7±4.77µs 26.0 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_1000_null_0.10       1.02   1361.7±5.03µs 45.9 MElem/sec      1.00   1340.7±3.19µs 46.6 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_20_null_0.10         1.00   1231.4±2.84µs 50.8 MElem/sec      1.01   1246.3±1.84µs 50.1 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_300_null_0.10        1.03   1276.4±8.24µs 49.0 MElem/sec      1.00   1236.9±2.77µs 50.5 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_65536_null_0.10      1.11      7.6±0.05ms  8.3 MElem/sec      1.00      6.8±0.07ms  9.1 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_75_null_0.10         1.01   1252.0±3.83µs 49.9 MElem/sec      1.00   1239.5±2.83µs 50.4 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_100000                    1.00     20.6±0.63ms 12.1 MElem/sec      1.01     20.9±0.47ms 12.0 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_32768                     1.00      7.8±0.02ms 32.2 MElem/sec      1.00      7.8±0.02ms 32.3 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_500000                    1.00     73.4±1.38ms  3.4 MElem/sec      1.04     76.4±1.76ms  3.3 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_65536                     1.00     14.1±0.18ms 17.8 MElem/sec      1.05     14.8±0.13ms 16.9 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_8192                      1.01      2.8±0.01ms 87.8 MElem/sec      1.00      2.8±0.02ms 88.4 MElem/sec
dict_take_n/take_n/size_65536_card_1000_null_0.10                                 1.00      2.5±0.01ms 101.0 MElem/sec     1.02      2.5±0.00ms 99.0 MElem/sec
dict_take_n/take_n/size_65536_card_20_null_0.10                                   1.00      2.2±0.01ms 116.0 MElem/sec     1.08      2.3±0.00ms 107.7 MElem/sec
dict_take_n/take_n/size_65536_card_300_null_0.10                                  1.00      2.3±0.01ms 109.8 MElem/sec     1.03      2.4±0.01ms 106.3 MElem/sec
dict_take_n/take_n/size_65536_card_65536_null_0.10                                1.03     27.8±0.13ms  9.0 MElem/sec      1.00     26.9±0.15ms  9.3 MElem/sec
dict_take_n/take_n/size_65536_card_75_null_0.10                                   1.00      2.2±0.01ms 112.7 MElem/sec     1.04      2.3±0.01ms 107.9 MElem/sec
group_values_emit/dictionary/size_500000_null_0.0                                 371.11     3.9±0.94ms 121.4 MElem/sec    1.00     10.6±1.23µs 44.0 GElem/sec
group_values_emit/dictionary/size_500000_null_0.1                                 577.90     8.9±0.66ms 53.7 MElem/sec     1.00     15.4±1.61µs 30.3 GElem/sec
group_values_emit/dictionary/size_65536_null_0.0                                  10.72    61.4±2.28µs 1018.6 MElem/sec    1.00      5.7±0.60µs 10.7 GElem/sec
group_values_emit/dictionary/size_65536_null_0.1                                  106.24  730.8±12.04µs 85.5 MElem/sec     1.00      6.9±0.46µs  8.9 GElem/sec
group_values_emit/dictionary/size_8192_null_0.0                                   6.52   64.3±526.68µs 121.5 MElem/sec     1.00     9.9±64.99µs 792.2 MElem/sec
group_values_emit/dictionary/size_8192_null_0.1                                   26.94   98.1±17.29µs 79.7 MElem/sec      1.00      3.6±8.54µs  2.1 GElem/sec
group_values_emit/utf8_view/size_500000_null_0.0                                  522.97     2.6±0.07ms 184.9 MElem/sec    1.00      4.9±0.51µs 94.4 GElem/sec
group_values_emit/utf8_view/size_500000_null_0.1                                  1.07     10.2±0.79µs 45.7 GElem/sec      1.00      9.5±0.75µs 48.9 GElem/sec
group_values_emit/utf8_view/size_65536_null_0.0                                   1.00  1802.7±188.39ns 33.9 GElem/sec     1.03  1853.2±258.34ns 32.9 GElem/sec
group_values_emit/utf8_view/size_65536_null_0.1                                   1.00      2.4±0.20µs 25.1 GElem/sec      1.12      2.7±0.18µs 22.4 GElem/sec
group_values_emit/utf8_view/size_8192_null_0.0                                    1.00  675.6±112.61ns 11.3 GElem/sec      1.03  698.9±113.22ns 10.9 GElem/sec
group_values_emit/utf8_view/size_8192_null_0.1                                    1.03  998.6±162.73ns  7.6 GElem/sec      1.00   966.6±97.93ns  7.9 GElem/sec

Resource Usage

dictionary_group_values — base (merge-base)

Metric	Value
Wall time	550.1s
Peak memory	886.3 MiB
Avg memory	112.4 MiB
CPU user	547.8s
CPU sys	42.5s
Peak spill	0 B

dictionary_group_values — branch

Metric	Value
Wall time	640.1s
Peak memory	902.7 MiB
Avg memory	208.9 MiB
CPU user	578.1s
CPU sys	23.7s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-11-03T16:21:39Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing b282594 (b282594) to 522287c diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "522287c20d3c09c1b2bd2a8a39bc3a164257614e"
changed:
  ref: "b2825948e531b1afe34769f29909359216175b50"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

group                                                                             HEAD                                     rich-T-kid_dictionary-aggregate-emission
-----                                                                             ----                                     ----------------------------------------
dict_intern_emit/intern_emit/size_65536_card_1000_null_0.00                       1.00    495.5±7.57µs 126.1 MElem/sec     1.01    500.2±6.25µs 125.0 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_20_null_0.00                         1.00    428.0±7.32µs 146.0 MElem/sec     1.03    442.9±5.90µs 141.1 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_300_null_0.00                        1.00    456.1±7.77µs 137.0 MElem/sec     1.01    461.6±6.65µs 135.4 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_65536_null_0.00                      1.00      6.3±0.09ms 10.0 MElem/sec      1.07      6.7±0.15ms  9.3 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_75_null_0.00                         1.00    444.4±7.44µs 140.6 MElem/sec     1.03    458.5±5.51µs 136.3 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_1000_null_0.00                        1.00    114.9±1.99µs 68.0 MElem/sec      1.01    115.9±1.86µs 67.4 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_20_null_0.00                          1.00     58.2±1.49µs 134.2 MElem/sec     1.01     58.7±1.39µs 133.2 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_300_null_0.00                         1.00     75.3±3.83µs 103.7 MElem/sec     1.00     75.4±1.90µs 103.6 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_75_null_0.00                          1.00     63.5±1.99µs 123.0 MElem/sec     1.01     64.0±1.57µs 122.1 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_8192_null_0.00                        1.00    527.9±3.40µs 14.8 MElem/sec      1.01    534.0±1.90µs 14.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_1000_null_0.10     1.00      2.4±0.01ms 105.3 MElem/sec     1.00      2.4±0.01ms 105.0 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_20_null_0.10       1.00      2.2±0.01ms 113.3 MElem/sec     1.06      2.3±0.01ms 107.3 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_300_null_0.10      1.00      2.3±0.01ms 109.1 MElem/sec     1.01      2.3±0.01ms 107.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_65536_null_0.10    1.18     23.9±0.61ms 10.5 MElem/sec      1.00     20.2±1.02ms 12.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_75_null_0.10       1.00      2.3±0.01ms 110.4 MElem/sec     1.03      2.3±0.01ms 106.9 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_1000_null_0.10      1.03    402.6±1.98µs 77.6 MElem/sec      1.00    390.9±1.80µs 79.9 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_20_null_0.10        1.00    266.0±1.69µs 117.5 MElem/sec     1.06    281.0±1.82µs 111.2 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_300_null_0.10       1.00    308.6±6.05µs 101.3 MElem/sec     1.00    307.2±1.91µs 101.7 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_75_null_0.10        1.00    279.6±2.13µs 111.8 MElem/sec     1.02    285.2±1.42µs 109.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_8192_null_0.10      1.09   1320.2±7.15µs 23.7 MElem/sec      1.00   1215.8±4.44µs 25.7 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_1000_null_0.10       1.05   1370.6±4.27µs 45.6 MElem/sec      1.00   1304.7±3.75µs 47.9 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_20_null_0.10         1.00   1234.6±4.08µs 50.6 MElem/sec      1.01   1248.5±3.45µs 50.1 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_300_null_0.10        1.03   1274.2±5.21µs 49.0 MElem/sec      1.00   1235.9±3.21µs 50.6 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_65536_null_0.10      1.22      8.4±0.16ms  7.4 MElem/sec      1.00      6.9±0.12ms  9.0 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_75_null_0.10         1.02   1291.6±4.74µs 48.4 MElem/sec      1.00   1267.8±3.29µs 49.3 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_100000                    1.29     26.4±0.89ms  9.5 MElem/sec      1.00     20.4±0.95ms 12.2 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_32768                     1.03      8.0±0.11ms 31.4 MElem/sec      1.00      7.7±0.05ms 32.5 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_500000                    1.11     94.9±2.19ms  2.6 MElem/sec      1.00     85.2±8.64ms  2.9 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_65536                     1.12     17.1±0.52ms 14.6 MElem/sec      1.00     15.2±0.91ms 16.4 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_8192                      1.03      2.9±0.02ms 85.6 MElem/sec      1.00      2.8±0.02ms 88.0 MElem/sec
dict_take_n/take_n/size_65536_card_1000_null_0.10                                 1.00      2.5±0.02ms 99.5 MElem/sec      1.02      2.6±0.01ms 97.7 MElem/sec
dict_take_n/take_n/size_65536_card_20_null_0.10                                   1.00      2.2±0.01ms 115.1 MElem/sec     1.08      2.3±0.01ms 106.7 MElem/sec
dict_take_n/take_n/size_65536_card_300_null_0.10                                  1.00      2.3±0.01ms 107.9 MElem/sec     1.02      2.4±0.01ms 105.6 MElem/sec
dict_take_n/take_n/size_65536_card_65536_null_0.10                                1.00     27.7±0.59ms  9.0 MElem/sec      1.00     27.7±0.52ms  9.0 MElem/sec
dict_take_n/take_n/size_65536_card_75_null_0.10                                   1.00      2.2±0.01ms 111.4 MElem/sec     1.03      2.3±0.01ms 108.1 MElem/sec
group_values_emit/dictionary/size_500000_null_0.0                                 298.85     3.4±0.19ms 139.5 MElem/sec    1.00     11.4±1.47µs 40.7 GElem/sec
group_values_emit/dictionary/size_500000_null_0.1                                 469.67     8.2±0.41ms 58.1 MElem/sec     1.00     17.5±1.96µs 26.7 GElem/sec
group_values_emit/dictionary/size_65536_null_0.0                                  12.41  80.1±123.15µs 780.1 MElem/sec     1.00      6.5±0.52µs  9.5 GElem/sec
group_values_emit/dictionary/size_65536_null_0.1                                  93.16  742.7±16.65µs 84.1 MElem/sec      1.00      8.0±0.71µs  7.7 GElem/sec
group_values_emit/dictionary/size_8192_null_0.0                                   1.12   33.1±186.83µs 235.9 MElem/sec     1.00   29.5±269.22µs 264.6 MElem/sec
group_values_emit/dictionary/size_8192_null_0.1                                   26.55    98.8±9.41µs 79.0 MElem/sec      1.00      3.7±6.12µs  2.0 GElem/sec
group_values_emit/utf8_view/size_500000_null_0.0                                  441.42     2.4±0.11ms 194.9 MElem/sec    1.00      5.5±0.70µs 84.0 GElem/sec
group_values_emit/utf8_view/size_500000_null_0.1                                  1.00     11.1±1.05µs 42.0 GElem/sec      1.12     12.5±1.03µs 37.4 GElem/sec
group_values_emit/utf8_view/size_65536_null_0.0                                   1.09      2.4±0.30µs 25.6 GElem/sec      1.00      2.2±0.28µs 27.9 GElem/sec
group_values_emit/utf8_view/size_65536_null_0.1                                   1.02      3.3±0.33µs 18.4 GElem/sec      1.00      3.3±0.34µs 18.7 GElem/sec
group_values_emit/utf8_view/size_8192_null_0.0                                    1.15  978.0±239.81ns  7.8 GElem/sec      1.00  852.2±156.20ns  9.0 GElem/sec
group_values_emit/utf8_view/size_8192_null_0.1                                    1.05  1227.4±233.57ns  6.2 GElem/sec     1.00  1165.5±181.34ns  6.5 GElem/sec

Resource Usage

dictionary_group_values — base (merge-base)

Metric	Value
Wall time	560.1s
Peak memory	900.7 MiB
Avg memory	111.4 MiB
CPU user	555.9s
CPU sys	41.7s
Peak spill	0 B

dictionary_group_values — branch

Metric	Value
Wall time	655.1s
Peak memory	902.3 MiB
Avg memory	211.7 MiB
CPU user	589.9s
CPU sys	21.7s
Peak spill	0 B

File an issue against this benchmark runner

Rich-T-kid · 2026-11-03T18:23:36Z

run benchmark dictionary_group_values

baseline:
ref: "2137bfabd7b4cd0ce1bbbc9095e58699fd3db643"
changed:
ref: "5ee33fb22e66766c549e7411998684decc6d072c"

Rich-T-kid · 2026-11-03T18:23:37Z

run benchmark dictionary_group_values

baseline:
ref: "2137bfabd7b4cd0ce1bbbc9095e58699fd3db643"
changed:
ref: "5ee33fb22e66766c549e7411998684decc6d072c"

adriangbot · 2026-11-04T03:31:42Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6621747758-4419-h9wj7 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing d0d542b (d0d542b) to ea33cce diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "d0d542bfd3f87bf266e7141b432bce2200066ae6"

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-11-04T03:32:15Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6621746703-4418-42xlk 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing d0d542b (d0d542b) to ea33cce diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "d0d542bfd3f87bf266e7141b432bce2200066ae6"

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-11-04T03:52:47Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing d0d542b (d0d542b) to ea33cce diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "d0d542bfd3f87bf266e7141b432bce2200066ae6"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

group                                                                             HEAD                                    rich-T-kid_dictionary-aggregate-emission
-----                                                                             ----                                    ----------------------------------------
dict_intern_emit/intern_emit/size_65536_card_1000_null_0.00                       1.03    489.0±7.60µs 127.8 MElem/sec    1.00    476.8±5.87µs 131.1 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_20_null_0.00                         1.02    424.9±7.42µs 147.1 MElem/sec    1.00    414.7±3.99µs 150.7 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_300_null_0.00                        1.03    450.7±7.18µs 138.7 MElem/sec    1.00    439.7±4.29µs 142.1 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_65536_null_0.00                      1.02      6.3±0.02ms 10.0 MElem/sec     1.00      6.2±0.01ms 10.2 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_75_null_0.00                         1.02    441.3±6.76µs 141.6 MElem/sec    1.00    433.5±4.49µs 144.2 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_1000_null_0.00                        1.03    113.3±1.40µs 69.0 MElem/sec     1.00    109.6±0.86µs 71.3 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_20_null_0.00                          1.05     57.4±1.31µs 136.1 MElem/sec    1.00     54.9±0.66µs 142.3 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_300_null_0.00                         1.03     73.2±1.28µs 106.7 MElem/sec    1.00     70.9±0.71µs 110.1 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_75_null_0.00                          1.00     61.5±1.22µs 127.0 MElem/sec    1.03    63.2±37.03µs 123.6 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_8192_null_0.00                        1.03    529.0±3.33µs 14.8 MElem/sec     1.00    511.5±1.56µs 15.3 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_1000_null_0.10     1.01      2.3±0.01ms 107.0 MElem/sec    1.00      2.3±0.00ms 107.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_20_null_0.10       1.00      2.2±0.00ms 116.0 MElem/sec    1.00      2.1±0.00ms 116.5 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_300_null_0.10      1.01      2.3±0.00ms 111.0 MElem/sec    1.00      2.2±0.00ms 111.9 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_65536_null_0.10    1.14     19.8±0.50ms 12.6 MElem/sec     1.00     17.4±0.11ms 14.3 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_75_null_0.10       1.00      2.2±0.00ms 111.9 MElem/sec    1.00      2.2±0.00ms 112.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_1000_null_0.10      1.04    398.2±2.66µs 78.5 MElem/sec     1.00    381.6±1.94µs 81.9 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_20_null_0.10        1.00    247.6±1.97µs 126.2 MElem/sec    1.00    247.8±1.33µs 126.1 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_300_null_0.10       1.03    296.5±4.08µs 105.4 MElem/sec    1.00    287.4±1.71µs 108.7 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_75_null_0.10        1.01    265.1±2.37µs 117.9 MElem/sec    1.00    262.4±1.24µs 119.1 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_8192_null_0.10      1.09   1318.4±7.92µs 23.7 MElem/sec     1.00   1206.8±4.36µs 25.9 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_1000_null_0.10       1.04   1358.1±3.90µs 46.0 MElem/sec     1.00   1304.1±4.28µs 47.9 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_20_null_0.10         1.05   1261.4±4.23µs 49.5 MElem/sec     1.00   1196.7±2.50µs 52.2 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_300_null_0.10        1.06   1302.7±5.44µs 48.0 MElem/sec     1.00   1225.2±1.24µs 51.0 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_65536_null_0.10      1.15      7.8±0.05ms  8.0 MElem/sec     1.00      6.8±0.02ms  9.2 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_75_null_0.10         1.03   1249.5±5.17µs 50.0 MElem/sec     1.00   1215.5±1.56µs 51.4 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_100000                    1.21     22.1±0.51ms 11.3 MElem/sec     1.00     18.2±0.21ms 13.7 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_32768                     1.06      8.0±0.02ms 31.1 MElem/sec     1.00      7.6±0.02ms 33.1 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_500000                    1.20     79.8±1.76ms  3.1 MElem/sec     1.00     66.6±1.05ms  3.8 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_65536                     1.07     15.2±0.23ms 16.4 MElem/sec     1.00     14.2±0.04ms 17.6 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_8192                      1.04      2.9±0.02ms 86.7 MElem/sec     1.00      2.8±0.01ms 90.1 MElem/sec
dict_take_n/take_n/size_65536_card_1000_null_0.10                                 1.00      2.5±0.01ms 101.7 MElem/sec    1.00      2.4±0.01ms 102.1 MElem/sec
dict_take_n/take_n/size_65536_card_20_null_0.10                                   1.00      2.1±0.01ms 120.7 MElem/sec    1.00      2.1±0.00ms 121.2 MElem/sec
dict_take_n/take_n/size_65536_card_300_null_0.10                                  1.01      2.2±0.01ms 111.8 MElem/sec    1.00      2.2±0.00ms 112.4 MElem/sec
dict_take_n/take_n/size_65536_card_65536_null_0.10                                1.07     27.9±0.15ms  8.9 MElem/sec     1.00     26.2±0.15ms  9.6 MElem/sec
dict_take_n/take_n/size_65536_card_75_null_0.10                                   1.00      2.2±0.01ms 115.0 MElem/sec    1.00      2.2±0.00ms 115.4 MElem/sec

Resource Usage

dictionary_group_values — base (merge-base)

Metric	Value
Wall time	600.1s
Peak memory	886.3 MiB
Avg memory	70.6 MiB
CPU user	415.3s
CPU sys	24.6s
Peak spill	0 B

dictionary_group_values — branch

Metric	Value
Wall time	620.1s
Peak memory	902.7 MiB
Avg memory	157.6 MiB
CPU user	424.3s
CPU sys	14.9s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-11-04T03:52:59Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing d0d542b (d0d542b) to ea33cce diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "d0d542bfd3f87bf266e7141b432bce2200066ae6"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

group                                                                             HEAD                                    rich-T-kid_dictionary-aggregate-emission
-----                                                                             ----                                    ----------------------------------------
dict_intern_emit/intern_emit/size_65536_card_1000_null_0.00                       1.00    486.6±6.93µs 128.5 MElem/sec    1.01    491.7±8.51µs 127.1 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_20_null_0.00                         1.01    421.6±9.09µs 148.2 MElem/sec    1.00    417.6±6.52µs 149.7 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_300_null_0.00                        1.02    451.1±7.50µs 138.5 MElem/sec    1.00    443.0±4.78µs 141.1 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_65536_null_0.00                      1.05      6.2±0.05ms 10.1 MElem/sec     1.00      5.9±0.03ms 10.6 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_75_null_0.00                         1.03    445.9±8.24µs 140.2 MElem/sec    1.00    433.9±4.52µs 144.0 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_1000_null_0.00                        1.03    113.8±2.05µs 68.7 MElem/sec     1.00    110.8±1.68µs 70.5 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_20_null_0.00                          1.02     57.1±1.40µs 136.9 MElem/sec    1.00     55.7±1.08µs 140.1 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_300_null_0.00                         1.00     72.7±1.10µs 107.5 MElem/sec    1.00     72.4±1.45µs 108.0 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_75_null_0.00                          1.01     62.6±1.54µs 124.8 MElem/sec    1.00    61.9±14.90µs 126.3 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_8192_null_0.00                        1.01    521.8±3.03µs 15.0 MElem/sec     1.00    516.5±2.04µs 15.1 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_1000_null_0.10     1.01      2.3±0.01ms 106.6 MElem/sec    1.00      2.3±0.00ms 107.5 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_20_null_0.10       1.00      2.2±0.01ms 116.1 MElem/sec    1.00      2.2±0.00ms 116.2 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_300_null_0.10      1.01      2.3±0.01ms 110.5 MElem/sec    1.00      2.2±0.01ms 111.2 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_65536_null_0.10    1.14     21.4±0.38ms 11.7 MElem/sec     1.00     18.8±0.59ms 13.3 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_75_null_0.10       1.00      2.2±0.01ms 111.9 MElem/sec    1.00      2.2±0.00ms 112.1 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_1000_null_0.10      1.04    397.0±3.37µs 78.7 MElem/sec     1.00    382.0±1.54µs 81.8 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_20_null_0.10        1.01    249.4±8.23µs 125.3 MElem/sec    1.00    247.2±1.49µs 126.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_300_null_0.10       1.03    296.2±2.80µs 105.5 MElem/sec    1.00    288.9±2.00µs 108.2 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_75_null_0.10        1.02   269.5±41.46µs 116.0 MElem/sec    1.00    263.1±1.58µs 118.8 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_8192_null_0.10      1.09  1316.9±67.51µs 23.7 MElem/sec     1.00   1206.9±5.25µs 25.9 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_1000_null_0.10       1.01   1357.3±3.88µs 46.0 MElem/sec     1.00   1341.5±5.40µs 46.6 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_20_null_0.10         1.02   1223.9±3.66µs 51.1 MElem/sec     1.00   1196.3±3.36µs 52.2 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_300_null_0.10        1.03   1266.0±4.85µs 49.4 MElem/sec     1.00   1229.9±3.30µs 50.8 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_65536_null_0.10      1.13      7.8±0.11ms  8.0 MElem/sec     1.00      6.9±0.03ms  9.0 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_75_null_0.10         1.03   1248.6±5.10µs 50.1 MElem/sec     1.00   1216.4±2.81µs 51.4 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_100000                    1.05     23.1±0.78ms 10.8 MElem/sec     1.00     21.9±0.87ms 11.4 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_32768                     1.06      8.2±0.03ms 30.4 MElem/sec     1.00      7.7±0.03ms 32.3 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_500000                    1.00     76.6±2.70ms  3.3 MElem/sec     1.06     81.2±2.65ms  3.1 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_65536                     1.04     15.8±1.00ms 15.8 MElem/sec     1.00     15.2±0.40ms 16.4 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_8192                      1.04      2.9±0.02ms 86.6 MElem/sec     1.00      2.8±0.01ms 90.4 MElem/sec
dict_take_n/take_n/size_65536_card_1000_null_0.10                                 1.00      2.5±0.01ms 101.9 MElem/sec    1.00      2.5±0.00ms 101.6 MElem/sec
dict_take_n/take_n/size_65536_card_20_null_0.10                                   1.00      2.1±0.01ms 120.7 MElem/sec    1.01      2.1±0.01ms 119.8 MElem/sec
dict_take_n/take_n/size_65536_card_300_null_0.10                                  1.00      2.3±0.01ms 111.1 MElem/sec    1.00      2.2±0.01ms 111.3 MElem/sec
dict_take_n/take_n/size_65536_card_65536_null_0.10                                1.02     27.9±0.14ms  9.0 MElem/sec     1.00     27.3±0.13ms  9.2 MElem/sec
dict_take_n/take_n/size_65536_card_75_null_0.10                                   1.00      2.2±0.01ms 115.2 MElem/sec    1.00      2.2±0.01ms 115.1 MElem/sec

Resource Usage

dictionary_group_values — base (merge-base)

Metric	Value
Wall time	650.2s
Peak memory	901.1 MiB
Avg memory	65.1 MiB
CPU user	415.9s
CPU sys	24.7s
Peak spill	0 B

dictionary_group_values — branch

Metric	Value
Wall time	635.2s
Peak memory	902.3 MiB
Avg memory	155.2 MiB
CPU user	428.8s
CPU sys	14.9s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-11-04T03:53:05Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing d0d542b (d0d542b) to ea33cce diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "d0d542bfd3f87bf266e7141b432bce2200066ae6"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

group                                                                             HEAD                                    rich-T-kid_dictionary-aggregate-emission
-----                                                                             ----                                    ----------------------------------------
dict_intern_emit/intern_emit/size_65536_card_1000_null_0.00                       1.00    484.4±6.94µs 129.0 MElem/sec    1.00    485.7±8.20µs 128.7 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_20_null_0.00                         1.00    417.9±6.19µs 149.6 MElem/sec    1.00    419.3±5.61µs 149.1 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_300_null_0.00                        1.00    450.0±6.68µs 138.9 MElem/sec    1.06  478.1±326.88µs 130.7 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_65536_null_0.00                      1.03      6.1±0.05ms 10.3 MElem/sec     1.00      5.9±0.02ms 10.7 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_75_null_0.00                         1.00    436.3±6.94µs 143.2 MElem/sec    1.01    439.3±5.76µs 142.3 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_1000_null_0.00                        1.02    112.6±1.24µs 69.4 MElem/sec     1.00    110.4±1.55µs 70.8 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_20_null_0.00                          1.03     57.2±1.16µs 136.7 MElem/sec    1.00     55.8±1.28µs 140.1 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_300_null_0.00                         1.02     73.1±1.14µs 106.9 MElem/sec    1.00     71.6±1.49µs 109.2 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_75_null_0.00                          1.03     61.9±1.30µs 126.3 MElem/sec    1.00     60.3±1.15µs 129.6 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_8192_null_0.00                        1.03    526.1±5.25µs 14.9 MElem/sec     1.00    511.7±1.66µs 15.3 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_1000_null_0.10     1.01      2.3±0.01ms 107.0 MElem/sec    1.00      2.3±0.00ms 107.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_20_null_0.10       1.00      2.2±0.01ms 116.1 MElem/sec    1.00      2.2±0.00ms 115.8 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_300_null_0.10      1.00      2.3±0.01ms 110.9 MElem/sec    1.00      2.3±0.00ms 111.0 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_65536_null_0.10    1.14     21.3±0.38ms 11.8 MElem/sec     1.00     18.7±0.46ms 13.3 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_75_null_0.10       1.00      2.2±0.01ms 111.8 MElem/sec    1.00      2.2±0.00ms 111.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_1000_null_0.10      1.04    397.2±2.16µs 78.7 MElem/sec     1.00    383.7±1.84µs 81.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_20_null_0.10        1.00    248.6±2.12µs 125.7 MElem/sec    1.00    249.6±1.15µs 125.2 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_300_null_0.10       1.02    293.9±3.77µs 106.3 MElem/sec    1.00    288.4±1.89µs 108.3 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_75_null_0.10        1.00    264.2±2.56µs 118.3 MElem/sec    1.00    263.1±1.43µs 118.8 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_8192_null_0.10      1.08   1313.7±7.97µs 23.8 MElem/sec     1.00   1211.6±5.32µs 25.8 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_1000_null_0.10       1.05   1362.9±5.20µs 45.9 MElem/sec     1.00   1301.7±3.09µs 48.0 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_20_null_0.10         1.06   1275.7±5.57µs 49.0 MElem/sec     1.00   1197.9±2.13µs 52.2 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_300_null_0.10        1.03   1263.5±4.54µs 49.5 MElem/sec     1.00   1230.8±2.51µs 50.8 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_65536_null_0.10      1.08      7.5±0.04ms  8.3 MElem/sec     1.00      7.0±0.04ms  9.0 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_75_null_0.10         1.03   1252.7±4.53µs 49.9 MElem/sec     1.00   1217.1±2.53µs 51.4 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_100000                    1.00     22.1±0.51ms 11.3 MElem/sec     1.06     23.5±0.52ms 10.7 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_32768                     1.01      7.9±0.03ms 31.8 MElem/sec     1.00      7.8±0.02ms 32.2 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_500000                    1.04     81.7±1.69ms  3.1 MElem/sec     1.00     78.4±2.16ms  3.2 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_65536                     1.00     15.0±0.30ms 16.6 MElem/sec     1.01     15.1±0.26ms 16.5 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_8192                      1.03      2.9±0.02ms 87.1 MElem/sec     1.00      2.8±0.01ms 89.9 MElem/sec
dict_take_n/take_n/size_65536_card_1000_null_0.10                                 1.00      2.5±0.01ms 101.6 MElem/sec    1.00      2.5±0.01ms 101.4 MElem/sec
dict_take_n/take_n/size_65536_card_20_null_0.10                                   1.00      2.1±0.01ms 120.5 MElem/sec    1.01      2.1±0.00ms 119.7 MElem/sec
dict_take_n/take_n/size_65536_card_300_null_0.10                                  1.00      2.2±0.01ms 112.1 MElem/sec    1.01      2.2±0.01ms 111.2 MElem/sec
dict_take_n/take_n/size_65536_card_65536_null_0.10                                1.00     26.9±0.17ms  9.3 MElem/sec     1.00     26.9±0.14ms  9.3 MElem/sec
dict_take_n/take_n/size_65536_card_75_null_0.10                                   1.00      2.2±0.00ms 115.5 MElem/sec    1.02      2.2±0.01ms 113.7 MElem/sec

Resource Usage

dictionary_group_values — base (merge-base)

Metric	Value
Wall time	630.1s
Peak memory	901.2 MiB
Avg memory	67.2 MiB
CPU user	413.9s
CPU sys	23.9s
Peak spill	0 B

dictionary_group_values — branch

Metric	Value
Wall time	640.1s
Peak memory	902.3 MiB
Avg memory	155.3 MiB
CPU user	432.3s
CPU sys	15.1s
Peak spill	0 B

File an issue against this benchmark runner

Rich-T-kid · 2026-11-04T03:58:45Z

run benchmark tpch

env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "d0d542bfd3f87bf266e7141b432bce2200066ae6"

Rich-T-kid · 2026-11-04T03:58:48Z

run benchmark tpcds

env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "d0d542bfd3f87bf266e7141b432bce2200066ae6"

Rich-T-kid · 2026-11-04T03:58:53Z

run benchmark clickbench_partitioned

env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "d0d542bfd3f87bf266e7141b432bce2200066ae6"

adriangbot · 2026-11-04T04:01:24Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6622151900-4430-n4k9b 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing d0d542b (d0d542b) to ea33cce diff

Run configuration

run benchmark tpcds
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "d0d542bfd3f87bf266e7141b432bce2200066ae6"

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-11-04T04:01:48Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6622152961-4431-22hbl 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing d0d542b (d0d542b) to ea33cce diff

Run configuration

run benchmark clickbench_partitioned
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "d0d542bfd3f87bf266e7141b432bce2200066ae6"

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-11-04T04:02:05Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6622151271-4429-5gdc4 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing d0d542b (d0d542b) to ea33cce diff

Run configuration

run benchmark tpch
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "d0d542bfd3f87bf266e7141b432bce2200066ae6"

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-11-04T04:10:36Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing d0d542b (d0d542b) to ea33cce diff

Run configuration

run benchmark tpch
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "d0d542bfd3f87bf266e7141b432bce2200066ae6"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and rich-T-kid_dictionary-aggregate-emission
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃      HEAD ┃ rich-T-kid_dictionary-aggregate-emission ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │  40.44 ms │                                 40.09 ms │ no change │
│ QQuery 2  │  19.24 ms │                                 19.36 ms │ no change │
│ QQuery 3  │  28.89 ms │                                 29.16 ms │ no change │
│ QQuery 4  │  17.93 ms │                                 17.83 ms │ no change │
│ QQuery 5  │  36.84 ms │                                 36.99 ms │ no change │
│ QQuery 6  │  14.21 ms │                                 14.11 ms │ no change │
│ QQuery 7  │  42.31 ms │                                 42.92 ms │ no change │
│ QQuery 8  │  41.44 ms │                                 41.76 ms │ no change │
│ QQuery 9  │  51.14 ms │                                 51.13 ms │ no change │
│ QQuery 10 │  60.10 ms │                                 60.49 ms │ no change │
│ QQuery 11 │  14.29 ms │                                 14.18 ms │ no change │
│ QQuery 12 │  20.83 ms │                                 21.18 ms │ no change │
│ QQuery 13 │  49.45 ms │                                 49.72 ms │ no change │
│ QQuery 14 │  24.48 ms │                                 24.49 ms │ no change │
│ QQuery 15 │  31.80 ms │                                 31.92 ms │ no change │
│ QQuery 16 │  13.62 ms │                                 13.58 ms │ no change │
│ QQuery 17 │  71.75 ms │                                 74.20 ms │ no change │
│ QQuery 18 │ 114.98 ms │                                117.29 ms │ no change │
│ QQuery 19 │  29.49 ms │                                 29.61 ms │ no change │
│ QQuery 20 │  34.46 ms │                                 34.77 ms │ no change │
│ QQuery 21 │  55.29 ms │                                 56.61 ms │ no change │
│ QQuery 22 │  16.23 ms │                                 16.33 ms │ no change │
└───────────┴───────────┴──────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                       ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                       │ 829.23ms │
│ Total Time (rich-T-kid_dictionary-aggregate-emission)   │ 837.72ms │
│ Average Time (HEAD)                                     │  37.69ms │
│ Average Time (rich-T-kid_dictionary-aggregate-emission) │  38.08ms │
│ Queries Faster                                          │        0 │
│ Queries Slower                                          │        0 │
│ Queries with No Change                                  │       22 │
│ Queries with Failure                                    │        0 │
└─────────────────────────────────────────────────────────┴──────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-T-kid_dictionary-aggregate-emission
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃                              HEAD ┃ rich-T-kid_dictionary-aggregate-emission ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │    40.44 / 41.74 ±1.23 / 43.31 ms │           40.09 / 40.78 ±1.06 / 42.88 ms │ no change │
│ QQuery 2  │    19.24 / 19.66 ±0.57 / 20.78 ms │           19.36 / 19.87 ±0.27 / 20.12 ms │ no change │
│ QQuery 3  │    28.89 / 30.43 ±1.35 / 32.30 ms │           29.16 / 29.96 ±0.81 / 31.45 ms │ no change │
│ QQuery 4  │    17.93 / 18.81 ±0.82 / 20.05 ms │           17.83 / 18.58 ±0.64 / 19.55 ms │ no change │
│ QQuery 5  │    36.84 / 37.13 ±0.34 / 37.81 ms │           36.99 / 37.21 ±0.18 / 37.43 ms │ no change │
│ QQuery 6  │    14.21 / 14.49 ±0.25 / 14.92 ms │           14.11 / 14.18 ±0.05 / 14.24 ms │ no change │
│ QQuery 7  │    42.31 / 43.90 ±1.98 / 47.68 ms │           42.92 / 44.08 ±0.80 / 45.40 ms │ no change │
│ QQuery 8  │    41.44 / 41.75 ±0.34 / 42.35 ms │           41.76 / 42.43 ±1.11 / 44.64 ms │ no change │
│ QQuery 9  │    51.14 / 52.54 ±0.87 / 53.60 ms │           51.13 / 52.65 ±0.79 / 53.48 ms │ no change │
│ QQuery 10 │    60.10 / 60.38 ±0.22 / 60.75 ms │           60.49 / 61.31 ±0.81 / 62.44 ms │ no change │
│ QQuery 11 │    14.29 / 14.41 ±0.10 / 14.58 ms │           14.18 / 14.38 ±0.12 / 14.53 ms │ no change │
│ QQuery 12 │    20.83 / 21.05 ±0.21 / 21.40 ms │           21.18 / 21.27 ±0.07 / 21.36 ms │ no change │
│ QQuery 13 │    49.45 / 50.38 ±0.78 / 51.42 ms │           49.72 / 50.19 ±0.38 / 50.52 ms │ no change │
│ QQuery 14 │    24.48 / 24.63 ±0.14 / 24.84 ms │           24.49 / 24.74 ±0.17 / 24.97 ms │ no change │
│ QQuery 15 │    31.80 / 32.05 ±0.23 / 32.47 ms │           31.92 / 32.97 ±0.92 / 34.10 ms │ no change │
│ QQuery 16 │    13.62 / 13.91 ±0.15 / 14.05 ms │           13.58 / 13.74 ±0.10 / 13.87 ms │ no change │
│ QQuery 17 │    71.75 / 73.71 ±1.62 / 76.20 ms │           74.20 / 74.86 ±0.61 / 75.83 ms │ no change │
│ QQuery 18 │ 114.98 / 120.33 ±4.27 / 127.20 ms │        117.29 / 120.27 ±3.72 / 127.53 ms │ no change │
│ QQuery 19 │    29.49 / 30.40 ±0.81 / 31.67 ms │           29.61 / 29.83 ±0.25 / 30.30 ms │ no change │
│ QQuery 20 │    34.46 / 34.70 ±0.31 / 35.29 ms │           34.77 / 35.11 ±0.43 / 35.90 ms │ no change │
│ QQuery 21 │    55.29 / 58.64 ±2.69 / 62.23 ms │           56.61 / 59.28 ±3.01 / 64.80 ms │ no change │
│ QQuery 22 │    16.23 / 16.82 ±0.75 / 18.30 ms │           16.33 / 16.65 ±0.26 / 16.92 ms │ no change │
└───────────┴───────────────────────────────────┴──────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                       ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                       │ 851.87ms │
│ Total Time (rich-T-kid_dictionary-aggregate-emission)   │ 854.33ms │
│ Average Time (HEAD)                                     │  38.72ms │
│ Average Time (rich-T-kid_dictionary-aggregate-emission) │  38.83ms │
│ Queries Faster                                          │        0 │
│ Queries Slower                                          │        0 │
│ Queries with No Change                                  │       22 │
│ Queries with Failure                                    │        0 │
└─────────────────────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric	Value
Wall time	5.0s
Peak memory	1.6 GiB
Avg memory	736.8 MiB
CPU user	21.5s
CPU sys	1.7s
Peak spill	0 B

tpch — branch

Metric	Value
Wall time	5.0s
Peak memory	2.0 GiB
Avg memory	802.0 MiB
CPU user	21.7s
CPU sys	1.6s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-11-04T04:11:59Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing d0d542b (d0d542b) to ea33cce diff

Run configuration

run benchmark tpcds
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "d0d542bfd3f87bf266e7141b432bce2200066ae6"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and rich-T-kid_dictionary-aggregate-emission
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ rich-T-kid_dictionary-aggregate-emission ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │    5.21 ms │                                  5.67 ms │  1.09x slower │
│ QQuery 2  │   75.59 ms │                                 77.16 ms │     no change │
│ QQuery 3  │   28.83 ms │                                 29.15 ms │     no change │
│ QQuery 4  │  837.03 ms │                                880.44 ms │  1.05x slower │
│ QQuery 5  │   51.57 ms │                                 50.84 ms │     no change │
│ QQuery 6  │   34.14 ms │                                 33.63 ms │     no change │
│ QQuery 7  │   71.81 ms │                                 71.15 ms │     no change │
│ QQuery 8  │   38.73 ms │                                 38.60 ms │     no change │
│ QQuery 9  │   52.53 ms │                                 54.84 ms │     no change │
│ QQuery 10 │   59.43 ms │                                 59.41 ms │     no change │
│ QQuery 11 │  531.26 ms │                                502.13 ms │ +1.06x faster │
│ QQuery 12 │   26.85 ms │                                 26.14 ms │     no change │
│ QQuery 13 │  113.92 ms │                                111.70 ms │     no change │
│ QQuery 14 │  384.06 ms │                                378.17 ms │     no change │
│ QQuery 15 │   54.46 ms │                                 53.45 ms │     no change │
│ QQuery 16 │    6.65 ms │                                  6.38 ms │     no change │
│ QQuery 17 │   79.11 ms │                                 78.07 ms │     no change │
│ QQuery 18 │  108.80 ms │                                104.24 ms │     no change │
│ QQuery 19 │   42.31 ms │                                 41.63 ms │     no change │
│ QQuery 20 │   31.96 ms │                                 31.20 ms │     no change │
│ QQuery 21 │   17.49 ms │                                 17.39 ms │     no change │
│ QQuery 22 │   75.68 ms │                                 70.55 ms │ +1.07x faster │
│ QQuery 23 │  300.65 ms │                                296.86 ms │     no change │
│ QQuery 24 │  199.13 ms │                                199.62 ms │     no change │
│ QQuery 25 │  104.99 ms │                                104.80 ms │     no change │
│ QQuery 26 │   45.36 ms │                                 45.02 ms │     no change │
│ QQuery 27 │    6.40 ms │                                  6.23 ms │     no change │
│ QQuery 28 │   57.74 ms │                                 56.55 ms │     no change │
│ QQuery 29 │   96.66 ms │                                 93.42 ms │     no change │
│ QQuery 30 │   36.47 ms │                                 37.09 ms │     no change │
│ QQuery 31 │  110.24 ms │                                111.77 ms │     no change │
│ QQuery 32 │   20.22 ms │                                 20.78 ms │     no change │
│ QQuery 33 │   37.69 ms │                                 39.12 ms │     no change │
│ QQuery 34 │    9.29 ms │                                  9.98 ms │  1.07x slower │
│ QQuery 35 │   69.61 ms │                                 73.69 ms │  1.06x slower │
│ QQuery 36 │    5.75 ms │                                  6.33 ms │  1.10x slower │
│ QQuery 37 │    7.51 ms │                                  7.85 ms │     no change │
│ QQuery 38 │   64.39 ms │                                 67.29 ms │     no change │
│ QQuery 39 │   72.96 ms │                                 74.95 ms │     no change │
│ QQuery 40 │   23.00 ms │                                 24.49 ms │  1.06x slower │
│ QQuery 41 │   13.93 ms │                                 14.40 ms │     no change │
│ QQuery 42 │   23.80 ms │                                 24.32 ms │     no change │
│ QQuery 43 │    4.86 ms │                                  5.25 ms │  1.08x slower │
│ QQuery 44 │    8.64 ms │                                  9.14 ms │  1.06x slower │
│ QQuery 45 │   36.14 ms │                                 37.25 ms │     no change │
│ QQuery 46 │   11.37 ms │                                 11.55 ms │     no change │
│ QQuery 47 │  209.51 ms │                                205.59 ms │     no change │
│ QQuery 48 │   91.53 ms │                                 91.17 ms │     no change │
│ QQuery 49 │   70.95 ms │                                 70.43 ms │     no change │
│ QQuery 50 │   57.54 ms │                                 57.68 ms │     no change │
│ QQuery 51 │   95.30 ms │                                 93.73 ms │     no change │
│ QQuery 52 │   24.64 ms │                                 24.64 ms │     no change │
│ QQuery 53 │   29.40 ms │                                 29.84 ms │     no change │
│ QQuery 54 │   53.94 ms │                                 54.24 ms │     no change │
│ QQuery 55 │   24.04 ms │                                 24.04 ms │     no change │
│ QQuery 56 │   39.23 ms │                                 40.04 ms │     no change │
│ QQuery 57 │  153.62 ms │                                155.35 ms │     no change │
│ QQuery 58 │  104.46 ms │                                105.42 ms │     no change │
│ QQuery 59 │  106.82 ms │                                107.81 ms │     no change │
│ QQuery 60 │   43.21 ms │                                 43.57 ms │     no change │
│ QQuery 61 │    9.46 ms │                                  9.50 ms │     no change │
│ QQuery 62 │   38.20 ms │                                 38.43 ms │     no change │
│ QQuery 63 │   29.43 ms │                                 29.40 ms │     no change │
│ QQuery 64 │  356.26 ms │                                361.76 ms │     no change │
│ QQuery 65 │   76.56 ms │                                 77.88 ms │     no change │
│ QQuery 66 │   71.78 ms │                                 71.65 ms │     no change │
│ QQuery 67 │  238.63 ms │                                232.14 ms │     no change │
│ QQuery 68 │   11.81 ms │                                 11.42 ms │     no change │
│ QQuery 69 │   53.87 ms │                                 54.14 ms │     no change │
│ QQuery 70 │   98.10 ms │                                 98.17 ms │     no change │
│ QQuery 71 │   35.05 ms │                                 35.02 ms │     no change │
│ QQuery 72 │ 1537.79 ms │                               1458.69 ms │ +1.05x faster │
│ QQuery 73 │    9.89 ms │                                  9.72 ms │     no change │
│ QQuery 74 │  272.37 ms │                                259.22 ms │     no change │
│ QQuery 75 │  140.39 ms │                                138.59 ms │     no change │
│ QQuery 76 │   35.73 ms │                                 34.42 ms │     no change │
│ QQuery 77 │   60.18 ms │                                 59.18 ms │     no change │
│ QQuery 78 │  167.51 ms │                                161.72 ms │     no change │
│ QQuery 79 │   68.95 ms │                                 66.80 ms │     no change │
│ QQuery 80 │   94.89 ms │                                 94.23 ms │     no change │
│ QQuery 81 │   29.05 ms │                                 28.59 ms │     no change │
│ QQuery 82 │   17.50 ms │                                 17.23 ms │     no change │
│ QQuery 83 │   33.22 ms │                                 32.63 ms │     no change │
│ QQuery 84 │   31.13 ms │                                 30.91 ms │     no change │
│ QQuery 85 │   96.76 ms │                                 95.17 ms │     no change │
│ QQuery 86 │   25.25 ms │                                 25.25 ms │     no change │
│ QQuery 87 │   66.84 ms │                                 65.49 ms │     no change │
│ QQuery 88 │   59.99 ms │                                 58.88 ms │     no change │
│ QQuery 89 │   36.33 ms │                                 35.82 ms │     no change │
│ QQuery 90 │   16.66 ms │                                 16.43 ms │     no change │
│ QQuery 91 │   39.78 ms │                                 38.80 ms │     no change │
│ QQuery 92 │   30.48 ms │                                 29.76 ms │     no change │
│ QQuery 93 │   51.62 ms │                                 49.01 ms │ +1.05x faster │
│ QQuery 94 │   40.16 ms │                                 38.83 ms │     no change │
│ QQuery 95 │   82.34 ms │                                 80.96 ms │     no change │
│ QQuery 96 │   24.41 ms │                                 23.88 ms │     no change │
│ QQuery 97 │   51.59 ms │                                 50.90 ms │     no change │
│ QQuery 98 │   39.33 ms │                                 38.50 ms │     no change │
│ QQuery 99 │   56.73 ms │                                 57.14 ms │     no change │
└───────────┴────────────┴──────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                       ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                       │ 9434.45ms │
│ Total Time (rich-T-kid_dictionary-aggregate-emission)   │ 9319.46ms │
│ Average Time (HEAD)                                     │   95.30ms │
│ Average Time (rich-T-kid_dictionary-aggregate-emission) │   94.14ms │
│ Queries Faster                                          │         4 │
│ Queries Slower                                          │         8 │
│ Queries with No Change                                  │        87 │
│ Queries with Failure                                    │         0 │
└─────────────────────────────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-T-kid_dictionary-aggregate-emission
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃ rich-T-kid_dictionary-aggregate-emission ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.21 / 5.99 ±1.20 / 8.37 ms │              5.67 / 6.24 ±1.00 / 8.24 ms │     no change │
│ QQuery 2  │        75.59 / 76.17 ±0.45 / 76.83 ms │           77.16 / 77.40 ±0.21 / 77.68 ms │     no change │
│ QQuery 3  │        28.83 / 28.95 ±0.13 / 29.18 ms │           29.15 / 29.50 ±0.31 / 29.90 ms │     no change │
│ QQuery 4  │    837.03 / 885.15 ±33.97 / 923.27 ms │        880.44 / 892.80 ±9.43 / 906.94 ms │     no change │
│ QQuery 5  │        51.57 / 52.11 ±0.45 / 52.68 ms │           50.84 / 51.63 ±1.01 / 53.57 ms │     no change │
│ QQuery 6  │        34.14 / 34.45 ±0.22 / 34.80 ms │           33.63 / 34.20 ±0.37 / 34.74 ms │     no change │
│ QQuery 7  │        71.81 / 72.37 ±0.49 / 73.19 ms │           71.15 / 71.81 ±0.40 / 72.31 ms │     no change │
│ QQuery 8  │        38.73 / 39.13 ±0.24 / 39.46 ms │           38.60 / 40.33 ±3.04 / 46.40 ms │     no change │
│ QQuery 9  │        52.53 / 56.09 ±4.06 / 63.93 ms │           54.84 / 56.15 ±1.07 / 58.01 ms │     no change │
│ QQuery 10 │        59.43 / 59.70 ±0.17 / 59.90 ms │           59.41 / 59.78 ±0.24 / 60.13 ms │     no change │
│ QQuery 11 │     531.26 / 536.37 ±2.83 / 539.71 ms │       502.13 / 516.57 ±12.77 / 533.66 ms │     no change │
│ QQuery 12 │        26.85 / 26.99 ±0.18 / 27.34 ms │           26.14 / 26.43 ±0.18 / 26.65 ms │     no change │
│ QQuery 13 │     113.92 / 114.48 ±0.62 / 115.69 ms │        111.70 / 112.51 ±0.50 / 113.28 ms │     no change │
│ QQuery 14 │     384.06 / 387.37 ±2.70 / 390.90 ms │        378.17 / 384.22 ±3.85 / 390.22 ms │     no change │
│ QQuery 15 │        54.46 / 57.41 ±2.49 / 61.67 ms │           53.45 / 55.78 ±1.98 / 59.38 ms │     no change │
│ QQuery 16 │           6.65 / 6.75 ±0.14 / 7.01 ms │              6.38 / 6.56 ±0.16 / 6.86 ms │     no change │
│ QQuery 17 │        79.11 / 80.42 ±1.64 / 83.58 ms │           78.07 / 79.64 ±1.79 / 82.65 ms │     no change │
│ QQuery 18 │     108.80 / 110.51 ±2.60 / 115.67 ms │        104.24 / 106.48 ±1.28 / 107.93 ms │     no change │
│ QQuery 19 │        42.31 / 42.97 ±0.48 / 43.66 ms │           41.63 / 42.61 ±1.15 / 44.79 ms │     no change │
│ QQuery 20 │        31.96 / 32.36 ±0.28 / 32.72 ms │           31.20 / 31.41 ±0.21 / 31.77 ms │     no change │
│ QQuery 21 │        17.49 / 17.69 ±0.12 / 17.88 ms │           17.39 / 17.60 ±0.19 / 17.95 ms │     no change │
│ QQuery 22 │        75.68 / 76.89 ±0.69 / 77.63 ms │           70.55 / 71.38 ±0.69 / 72.59 ms │ +1.08x faster │
│ QQuery 23 │     300.65 / 311.82 ±8.28 / 325.26 ms │        296.86 / 299.80 ±2.17 / 303.38 ms │     no change │
│ QQuery 24 │     199.13 / 203.22 ±3.45 / 206.96 ms │        199.62 / 201.52 ±1.57 / 203.72 ms │     no change │
│ QQuery 25 │     104.99 / 107.34 ±1.59 / 109.92 ms │        104.80 / 106.75 ±1.39 / 108.32 ms │     no change │
│ QQuery 26 │        45.36 / 45.61 ±0.18 / 45.92 ms │           45.02 / 45.14 ±0.11 / 45.28 ms │     no change │
│ QQuery 27 │           6.40 / 6.58 ±0.20 / 6.91 ms │              6.23 / 6.39 ±0.10 / 6.49 ms │     no change │
│ QQuery 28 │        57.74 / 61.11 ±1.71 / 62.36 ms │           56.55 / 58.76 ±2.43 / 61.86 ms │     no change │
│ QQuery 29 │       96.66 / 99.59 ±1.90 / 102.57 ms │          93.42 / 96.84 ±5.25 / 107.26 ms │     no change │
│ QQuery 30 │        36.47 / 37.47 ±0.68 / 38.53 ms │           37.09 / 37.89 ±0.85 / 39.51 ms │     no change │
│ QQuery 31 │     110.24 / 112.15 ±2.75 / 117.55 ms │        111.77 / 112.38 ±0.68 / 113.59 ms │     no change │
│ QQuery 32 │        20.22 / 20.36 ±0.20 / 20.75 ms │           20.78 / 22.09 ±1.61 / 25.11 ms │  1.09x slower │
│ QQuery 33 │        37.69 / 38.05 ±0.22 / 38.38 ms │           39.12 / 39.77 ±0.51 / 40.61 ms │     no change │
│ QQuery 34 │           9.29 / 9.48 ±0.17 / 9.79 ms │            9.98 / 10.15 ±0.21 / 10.55 ms │  1.07x slower │
│ QQuery 35 │        69.61 / 69.84 ±0.20 / 70.14 ms │           73.69 / 74.57 ±0.80 / 75.99 ms │  1.07x slower │
│ QQuery 36 │           5.75 / 5.85 ±0.18 / 6.22 ms │              6.33 / 6.45 ±0.10 / 6.59 ms │  1.10x slower │
│ QQuery 37 │           7.51 / 7.60 ±0.09 / 7.74 ms │              7.85 / 8.03 ±0.14 / 8.27 ms │  1.06x slower │
│ QQuery 38 │        64.39 / 65.68 ±1.68 / 69.01 ms │           67.29 / 68.63 ±1.43 / 70.71 ms │     no change │
│ QQuery 39 │        72.96 / 74.40 ±1.48 / 76.27 ms │           74.95 / 75.75 ±0.48 / 76.32 ms │     no change │
│ QQuery 40 │        23.00 / 23.33 ±0.21 / 23.59 ms │           24.49 / 25.03 ±0.35 / 25.50 ms │  1.07x slower │
│ QQuery 41 │        13.93 / 14.02 ±0.07 / 14.09 ms │           14.40 / 14.60 ±0.15 / 14.87 ms │     no change │
│ QQuery 42 │        23.80 / 24.34 ±0.40 / 25.01 ms │           24.32 / 24.61 ±0.16 / 24.82 ms │     no change │
│ QQuery 43 │           4.86 / 4.94 ±0.14 / 5.22 ms │              5.25 / 5.37 ±0.10 / 5.53 ms │  1.09x slower │
│ QQuery 44 │           8.64 / 8.76 ±0.09 / 8.85 ms │              9.14 / 9.27 ±0.13 / 9.45 ms │  1.06x slower │
│ QQuery 45 │        36.14 / 37.80 ±2.13 / 41.84 ms │           37.25 / 38.59 ±2.09 / 42.71 ms │     no change │
│ QQuery 46 │        11.37 / 11.53 ±0.13 / 11.75 ms │           11.55 / 11.82 ±0.20 / 12.13 ms │     no change │
│ QQuery 47 │     209.51 / 214.94 ±3.62 / 219.82 ms │        205.59 / 210.37 ±3.56 / 215.62 ms │     no change │
│ QQuery 48 │        91.53 / 92.10 ±0.73 / 93.54 ms │           91.17 / 92.11 ±1.31 / 94.71 ms │     no change │
│ QQuery 49 │        70.95 / 72.89 ±2.20 / 77.17 ms │           70.43 / 71.14 ±0.73 / 72.54 ms │     no change │
│ QQuery 50 │        57.54 / 59.26 ±2.46 / 64.13 ms │           57.68 / 57.98 ±0.46 / 58.88 ms │     no change │
│ QQuery 51 │        95.30 / 96.72 ±1.08 / 98.49 ms │           93.73 / 96.05 ±1.65 / 98.12 ms │     no change │
│ QQuery 52 │        24.64 / 26.67 ±3.39 / 33.43 ms │           24.64 / 24.89 ±0.22 / 25.26 ms │ +1.07x faster │
│ QQuery 53 │        29.40 / 29.90 ±0.53 / 30.83 ms │           29.84 / 31.03 ±2.05 / 35.12 ms │     no change │
│ QQuery 54 │        53.94 / 55.29 ±1.93 / 59.04 ms │           54.24 / 55.20 ±0.57 / 55.71 ms │     no change │
│ QQuery 55 │        24.04 / 24.34 ±0.22 / 24.66 ms │           24.04 / 24.82 ±0.72 / 26.14 ms │     no change │
│ QQuery 56 │        39.23 / 39.44 ±0.14 / 39.58 ms │           40.04 / 40.39 ±0.41 / 41.18 ms │     no change │
│ QQuery 57 │     153.62 / 155.46 ±1.99 / 159.09 ms │        155.35 / 157.15 ±1.45 / 159.78 ms │     no change │
│ QQuery 58 │     104.46 / 105.70 ±2.03 / 109.74 ms │        105.42 / 105.77 ±0.55 / 106.87 ms │     no change │
│ QQuery 59 │     106.82 / 107.67 ±1.09 / 109.77 ms │        107.81 / 108.50 ±0.61 / 109.31 ms │     no change │
│ QQuery 60 │        43.21 / 43.78 ±0.59 / 44.86 ms │           43.57 / 44.14 ±0.61 / 45.05 ms │     no change │
│ QQuery 61 │           9.46 / 9.61 ±0.13 / 9.84 ms │              9.50 / 9.61 ±0.09 / 9.75 ms │     no change │
│ QQuery 62 │        38.20 / 40.01 ±2.23 / 44.38 ms │           38.43 / 38.77 ±0.28 / 39.22 ms │     no change │
│ QQuery 63 │        29.43 / 30.15 ±0.48 / 30.71 ms │           29.40 / 30.99 ±2.74 / 36.46 ms │     no change │
│ QQuery 64 │     356.26 / 360.72 ±4.13 / 366.83 ms │        361.76 / 365.10 ±3.40 / 370.07 ms │     no change │
│ QQuery 65 │        76.56 / 79.41 ±2.76 / 84.50 ms │           77.88 / 79.65 ±3.12 / 85.86 ms │     no change │
│ QQuery 66 │        71.78 / 72.18 ±0.26 / 72.56 ms │           71.65 / 74.84 ±2.57 / 79.22 ms │     no change │
│ QQuery 67 │     238.63 / 241.11 ±1.99 / 243.86 ms │        232.14 / 234.06 ±1.83 / 237.20 ms │     no change │
│ QQuery 68 │        11.81 / 11.88 ±0.06 / 11.98 ms │           11.42 / 11.53 ±0.14 / 11.79 ms │     no change │
│ QQuery 69 │        53.87 / 56.60 ±3.51 / 62.86 ms │           54.14 / 54.50 ±0.50 / 55.48 ms │     no change │
│ QQuery 70 │       98.10 / 99.72 ±1.84 / 102.53 ms │          98.17 / 99.23 ±0.73 / 100.25 ms │     no change │
│ QQuery 71 │        35.05 / 35.52 ±0.28 / 35.82 ms │           35.02 / 35.81 ±0.55 / 36.74 ms │     no change │
│ QQuery 72 │ 1537.79 / 1630.04 ±78.46 / 1737.82 ms │    1458.69 / 1525.73 ±91.42 / 1705.05 ms │ +1.07x faster │
│ QQuery 73 │         9.89 / 10.17 ±0.23 / 10.48 ms │              9.72 / 9.85 ±0.09 / 9.97 ms │     no change │
│ QQuery 74 │     272.37 / 280.55 ±7.15 / 292.84 ms │        259.22 / 266.33 ±5.60 / 274.75 ms │ +1.05x faster │
│ QQuery 75 │     140.39 / 142.75 ±2.99 / 148.46 ms │        138.59 / 140.34 ±2.50 / 145.26 ms │     no change │
│ QQuery 76 │        35.73 / 36.83 ±1.20 / 38.54 ms │           34.42 / 34.66 ±0.32 / 35.27 ms │ +1.06x faster │
│ QQuery 77 │        60.18 / 60.63 ±0.36 / 61.27 ms │           59.18 / 59.42 ±0.17 / 59.68 ms │     no change │
│ QQuery 78 │     167.51 / 171.79 ±3.19 / 175.59 ms │        161.72 / 166.91 ±5.41 / 177.24 ms │     no change │
│ QQuery 79 │        68.95 / 70.16 ±0.70 / 71.13 ms │           66.80 / 70.18 ±3.67 / 76.71 ms │     no change │
│ QQuery 80 │        94.89 / 97.10 ±1.37 / 98.72 ms │           94.23 / 95.33 ±1.29 / 97.86 ms │     no change │
│ QQuery 81 │        29.05 / 29.23 ±0.13 / 29.41 ms │           28.59 / 28.77 ±0.16 / 29.05 ms │     no change │
│ QQuery 82 │        17.50 / 17.87 ±0.20 / 18.05 ms │           17.23 / 17.41 ±0.17 / 17.63 ms │     no change │
│ QQuery 83 │        33.22 / 33.37 ±0.15 / 33.60 ms │           32.63 / 33.89 ±2.03 / 37.93 ms │     no change │
│ QQuery 84 │        31.13 / 31.50 ±0.34 / 32.03 ms │           30.91 / 31.67 ±0.81 / 32.93 ms │     no change │
│ QQuery 85 │       96.76 / 99.40 ±2.70 / 104.09 ms │           95.17 / 96.33 ±0.67 / 97.16 ms │     no change │
│ QQuery 86 │        25.25 / 25.97 ±0.50 / 26.78 ms │           25.25 / 25.53 ±0.26 / 26.01 ms │     no change │
│ QQuery 87 │        66.84 / 67.49 ±0.46 / 68.08 ms │           65.49 / 67.28 ±1.51 / 69.38 ms │     no change │
│ QQuery 88 │        59.99 / 62.31 ±3.02 / 67.72 ms │           58.88 / 59.38 ±0.31 / 59.65 ms │     no change │
│ QQuery 89 │        36.33 / 36.82 ±0.49 / 37.73 ms │           35.82 / 36.43 ±0.82 / 38.04 ms │     no change │
│ QQuery 90 │        16.66 / 16.79 ±0.07 / 16.86 ms │           16.43 / 16.56 ±0.11 / 16.72 ms │     no change │
│ QQuery 91 │        39.78 / 40.37 ±0.42 / 40.96 ms │           38.80 / 39.47 ±0.36 / 39.87 ms │     no change │
│ QQuery 92 │        30.48 / 31.11 ±0.57 / 32.12 ms │           29.76 / 30.74 ±1.28 / 33.22 ms │     no change │
│ QQuery 93 │        51.62 / 52.83 ±1.42 / 55.56 ms │           49.01 / 50.55 ±0.87 / 51.53 ms │     no change │
│ QQuery 94 │        40.16 / 41.44 ±1.24 / 43.74 ms │           38.83 / 39.59 ±0.88 / 41.31 ms │     no change │
│ QQuery 95 │        82.34 / 83.50 ±1.04 / 85.24 ms │           80.96 / 81.43 ±0.32 / 81.90 ms │     no change │
│ QQuery 96 │        24.41 / 24.57 ±0.12 / 24.74 ms │           23.88 / 24.98 ±1.60 / 28.16 ms │     no change │
│ QQuery 97 │        51.59 / 53.64 ±1.95 / 56.31 ms │           50.90 / 52.68 ±2.00 / 56.38 ms │     no change │
│ QQuery 98 │        39.33 / 39.65 ±0.33 / 40.20 ms │           38.50 / 38.79 ±0.26 / 39.25 ms │     no change │
│ QQuery 99 │        56.73 / 57.76 ±0.73 / 58.67 ms │           57.14 / 57.90 ±0.86 / 59.58 ms │     no change │
└───────────┴───────────────────────────────────────┴──────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                       ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                       │ 9707.92ms │
│ Total Time (rich-T-kid_dictionary-aggregate-emission)   │ 9523.00ms │
│ Average Time (HEAD)                                     │   98.06ms │
│ Average Time (rich-T-kid_dictionary-aggregate-emission) │   96.19ms │
│ Queries Faster                                          │         5 │
│ Queries Slower                                          │         8 │
│ Queries with No Change                                  │        86 │
│ Queries with Failure                                    │         0 │
└─────────────────────────────────────────────────────────┴───────────┘

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	50.0s
Peak memory	1.9 GiB
Avg memory	1.4 GiB
CPU user	189.7s
CPU sys	5.8s
Peak spill	0 B

tpcds — branch

Metric	Value
Wall time	50.0s
Peak memory	1.9 GiB
Avg memory	1.4 GiB
CPU user	184.9s
CPU sys	5.6s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-11-04T04:25:49Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing d0d542b (d0d542b) to ea33cce diff

Run configuration

run benchmark clickbench_partitioned
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "d0d542bfd3f87bf266e7141b432bce2200066ae6"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and rich-T-kid_dictionary-aggregate-emission
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃        HEAD ┃ rich-T-kid_dictionary-aggregate-emission ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │     1.09 ms │                                  1.14 ms │     no change │
│ QQuery 1  │    10.95 ms │                                 11.73 ms │  1.07x slower │
│ QQuery 2  │    34.49 ms │                                 35.76 ms │     no change │
│ QQuery 3  │    30.70 ms │                                 34.38 ms │  1.12x slower │
│ QQuery 4  │   235.04 ms │                                236.12 ms │     no change │
│ QQuery 5  │  1572.43 ms │                               1540.60 ms │     no change │
│ QQuery 6  │     1.13 ms │                                  1.20 ms │  1.06x slower │
│ QQuery 7  │    12.46 ms │                                 13.16 ms │  1.06x slower │
│ QQuery 8  │   339.31 ms │                                341.49 ms │     no change │
│ QQuery 9  │   488.11 ms │                                492.15 ms │     no change │
│ QQuery 10 │   200.96 ms │                                200.08 ms │     no change │
│ QQuery 11 │   212.13 ms │                                210.60 ms │     no change │
│ QQuery 12 │  1519.66 ms │                               1562.62 ms │     no change │
│ QQuery 13 │  2396.67 ms │                               2102.93 ms │ +1.14x faster │
│ QQuery 14 │  1818.05 ms │                               1827.01 ms │     no change │
│ QQuery 15 │   295.22 ms │                                291.98 ms │     no change │
│ QQuery 16 │  3649.67 ms │                               3762.57 ms │     no change │
│ QQuery 17 │  3738.12 ms │                               3861.44 ms │     no change │
│ QQuery 18 │  7120.29 ms │                               6796.04 ms │     no change │
│ QQuery 19 │    27.49 ms │                                 26.91 ms │     no change │
│ QQuery 20 │  1063.34 ms │                               1060.03 ms │     no change │
│ QQuery 21 │  1378.68 ms │                               1381.49 ms │     no change │
│ QQuery 22 │  3324.79 ms │                               3265.28 ms │     no change │
│ QQuery 23 │  9433.19 ms │                               9466.70 ms │     no change │
│ QQuery 24 │    97.29 ms │                                 98.24 ms │     no change │
│ QQuery 25 │   352.94 ms │                                357.13 ms │     no change │
│ QQuery 26 │    98.14 ms │                                100.60 ms │     no change │
│ QQuery 27 │  1190.90 ms │                               1156.99 ms │     no change │
│ QQuery 28 │  2907.23 ms │                               2920.49 ms │     no change │
│ QQuery 29 │    37.67 ms │                                 37.37 ms │     no change │
│ QQuery 30 │   550.87 ms │                                577.63 ms │     no change │
│ QQuery 31 │   549.81 ms │                                553.48 ms │     no change │
│ QQuery 32 │  1119.94 ms │                               1047.97 ms │ +1.07x faster │
│ QQuery 33 │ 14200.07 ms │                              14182.99 ms │     no change │
│ QQuery 34 │ 14870.85 ms │                              14653.40 ms │     no change │
│ QQuery 35 │   307.78 ms │                                287.29 ms │ +1.07x faster │
│ QQuery 36 │   214.67 ms │                                177.27 ms │ +1.21x faster │
│ QQuery 37 │    96.91 ms │                                 96.06 ms │     no change │
│ QQuery 38 │    84.83 ms │                                 79.04 ms │ +1.07x faster │
│ QQuery 39 │   417.80 ms │                                396.13 ms │ +1.05x faster │
│ QQuery 40 │    15.42 ms │                                 14.74 ms │     no change │
│ QQuery 41 │    14.29 ms │                                 13.19 ms │ +1.08x faster │
│ QQuery 42 │    13.57 ms │                                 13.20 ms │     no change │
└───────────┴─────────────┴──────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                       ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                       │ 76044.96ms │
│ Total Time (rich-T-kid_dictionary-aggregate-emission)   │ 75286.63ms │
│ Average Time (HEAD)                                     │  1768.49ms │
│ Average Time (rich-T-kid_dictionary-aggregate-emission) │  1750.85ms │
│ Queries Faster                                          │          7 │
│ Queries Slower                                          │          4 │
│ Queries with No Change                                  │         32 │
│ Queries with Failure                                    │          0 │
└─────────────────────────────────────────────────────────┴────────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-T-kid_dictionary-aggregate-emission
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                      HEAD ┃   rich-T-kid_dictionary-aggregate-emission ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │               1.09 / 1.69 ±1.13 / 3.96 ms │                1.14 / 1.74 ±1.13 / 4.00 ms │     no change │
│ QQuery 1  │            10.95 / 11.16 ±0.17 / 11.41 ms │             11.73 / 12.22 ±0.61 / 13.39 ms │  1.09x slower │
│ QQuery 2  │            34.49 / 34.78 ±0.36 / 35.47 ms │             35.76 / 36.72 ±0.90 / 38.22 ms │  1.06x slower │
│ QQuery 3  │            30.70 / 31.23 ±0.56 / 32.28 ms │             34.38 / 35.29 ±0.73 / 36.51 ms │  1.13x slower │
│ QQuery 4  │        235.04 / 248.66 ±11.04 / 267.87 ms │         236.12 / 260.49 ±13.66 / 277.92 ms │     no change │
│ QQuery 5  │     1572.43 / 1647.98 ±71.68 / 1770.81 ms │      1540.60 / 1608.12 ±92.33 / 1790.30 ms │     no change │
│ QQuery 6  │               1.13 / 1.30 ±0.25 / 1.79 ms │                1.20 / 1.37 ±0.25 / 1.87 ms │  1.06x slower │
│ QQuery 7  │            12.46 / 12.80 ±0.23 / 13.14 ms │             13.16 / 13.23 ±0.06 / 13.33 ms │     no change │
│ QQuery 8  │         339.31 / 347.70 ±9.21 / 361.45 ms │         341.49 / 356.84 ±13.61 / 379.43 ms │     no change │
│ QQuery 9  │        488.11 / 507.33 ±18.18 / 532.55 ms │         492.15 / 506.64 ±15.60 / 530.69 ms │     no change │
│ QQuery 10 │         200.96 / 206.11 ±7.49 / 221.01 ms │          200.08 / 204.64 ±4.87 / 213.56 ms │     no change │
│ QQuery 11 │         212.13 / 213.38 ±1.09 / 215.41 ms │          210.60 / 211.87 ±1.48 / 214.65 ms │     no change │
│ QQuery 12 │     1519.66 / 1580.35 ±40.81 / 1630.05 ms │      1562.62 / 1588.06 ±23.88 / 1620.07 ms │     no change │
│ QQuery 13 │    2396.67 / 2832.96 ±330.35 / 3242.98 ms │     2102.93 / 2624.68 ±296.83 / 2980.29 ms │ +1.08x faster │
│ QQuery 14 │     1818.05 / 1883.76 ±57.73 / 1983.72 ms │      1827.01 / 1879.98 ±51.41 / 1946.62 ms │     no change │
│ QQuery 15 │        295.22 / 320.41 ±23.42 / 353.25 ms │         291.98 / 334.62 ±28.14 / 360.78 ms │     no change │
│ QQuery 16 │    3649.67 / 3934.96 ±187.61 / 4212.65 ms │      3762.57 / 3941.57 ±98.02 / 4035.64 ms │     no change │
│ QQuery 17 │    3738.12 / 3982.61 ±191.58 / 4242.86 ms │     3861.44 / 3955.71 ±129.13 / 4203.77 ms │     no change │
│ QQuery 18 │    7120.29 / 7393.60 ±215.77 / 7762.23 ms │     6796.04 / 7299.34 ±300.37 / 7710.93 ms │     no change │
│ QQuery 19 │           27.49 / 35.51 ±14.69 / 64.83 ms │             26.91 / 30.75 ±6.02 / 42.63 ms │ +1.15x faster │
│ QQuery 20 │     1063.34 / 1083.44 ±11.08 / 1096.89 ms │      1060.03 / 1081.95 ±18.63 / 1109.72 ms │     no change │
│ QQuery 21 │     1378.68 / 1404.50 ±22.66 / 1433.08 ms │      1381.49 / 1397.60 ±13.32 / 1415.86 ms │     no change │
│ QQuery 22 │     3324.79 / 3400.56 ±74.48 / 3533.26 ms │      3265.28 / 3292.04 ±36.36 / 3358.49 ms │     no change │
│ QQuery 23 │     9433.19 / 9544.36 ±85.10 / 9675.80 ms │    9466.70 / 9767.78 ±197.98 / 10022.18 ms │     no change │
│ QQuery 24 │         97.29 / 136.78 ±41.09 / 206.10 ms │          98.24 / 124.87 ±35.51 / 192.54 ms │ +1.10x faster │
│ QQuery 25 │         352.94 / 361.73 ±5.72 / 369.40 ms │         357.13 / 366.40 ±10.08 / 385.98 ms │     no change │
│ QQuery 26 │          98.14 / 101.21 ±3.10 / 106.04 ms │          100.60 / 103.11 ±2.58 / 107.40 ms │     no change │
│ QQuery 27 │     1190.90 / 1217.60 ±23.18 / 1250.84 ms │      1156.99 / 1197.13 ±31.68 / 1244.81 ms │     no change │
│ QQuery 28 │     2907.23 / 2950.98 ±49.03 / 3032.34 ms │      2920.49 / 3030.11 ±61.25 / 3109.67 ms │     no change │
│ QQuery 29 │          37.67 / 62.39 ±38.27 / 137.06 ms │             37.37 / 42.39 ±8.42 / 59.18 ms │ +1.47x faster │
│ QQuery 30 │         550.87 / 560.32 ±8.40 / 572.40 ms │          577.63 / 583.52 ±6.22 / 592.50 ms │     no change │
│ QQuery 31 │        549.81 / 568.31 ±12.67 / 582.79 ms │         553.48 / 578.63 ±15.68 / 596.95 ms │     no change │
│ QQuery 32 │     1119.94 / 1178.68 ±61.04 / 1283.03 ms │      1047.97 / 1175.02 ±82.78 / 1301.85 ms │     no change │
│ QQuery 33 │ 14200.07 / 15432.26 ±852.29 / 16429.87 ms │ 14182.99 / 15142.01 ±1226.24 / 17454.49 ms │     no change │
│ QQuery 34 │ 14870.85 / 15490.20 ±530.07 / 16208.24 ms │  14653.40 / 15583.54 ±733.94 / 16531.74 ms │     no change │
│ QQuery 35 │        307.78 / 379.89 ±59.14 / 486.62 ms │         287.29 / 338.65 ±64.20 / 463.99 ms │ +1.12x faster │
│ QQuery 36 │         214.67 / 222.46 ±6.51 / 234.14 ms │         177.27 / 196.82 ±14.91 / 221.06 ms │ +1.13x faster │
│ QQuery 37 │          96.91 / 100.73 ±3.82 / 106.19 ms │           96.06 / 105.41 ±7.61 / 117.34 ms │     no change │
│ QQuery 38 │            84.83 / 90.74 ±4.54 / 97.53 ms │             79.04 / 83.27 ±4.86 / 92.35 ms │ +1.09x faster │
│ QQuery 39 │        417.80 / 435.49 ±11.48 / 453.06 ms │         396.13 / 414.08 ±12.52 / 432.68 ms │     no change │
│ QQuery 40 │            15.42 / 20.43 ±9.09 / 38.59 ms │             14.74 / 16.27 ±2.59 / 21.44 ms │ +1.26x faster │
│ QQuery 41 │            14.29 / 14.58 ±0.17 / 14.83 ms │             13.19 / 13.40 ±0.16 / 13.68 ms │ +1.09x faster │
│ QQuery 42 │            13.57 / 14.13 ±0.36 / 14.69 ms │             13.20 / 14.42 ±2.33 / 19.08 ms │     no change │
└───────────┴───────────────────────────────────────────┴────────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                       ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                       │ 80000.07ms │
│ Total Time (rich-T-kid_dictionary-aggregate-emission)   │ 79552.28ms │
│ Average Time (HEAD)                                     │  1860.47ms │
│ Average Time (rich-T-kid_dictionary-aggregate-emission) │  1850.05ms │
│ Queries Faster                                          │          9 │
│ Queries Slower                                          │          4 │
│ Queries with No Change                                  │         30 │
│ Queries with Failure                                    │          0 │
└─────────────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	405.1s
Peak memory	13.0 GiB
Avg memory	6.0 GiB
CPU user	3422.7s
CPU sys	226.8s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	400.1s
Peak memory	13.0 GiB
Avg memory	6.0 GiB
CPU user	3405.5s
CPU sys	219.4s
Peak spill	0 B

File an issue against this benchmark runner

Every group key comes from an already validated inner slot. Check the key width when allocating that slot so later key writes stay infallible and out of the per-row append path.

Rich-T-kid · 2026-11-04T16:41:10Z

run benchmark dictionary_group_values

baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "9d3137dc3fcc4fbf7d0dccdbc506553a2ffb702d"

Rich-T-kid · 2026-11-04T16:41:12Z

run benchmark tpch

env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "9d3137dc3fcc4fbf7d0dccdbc506553a2ffb702d"

Rich-T-kid · 2026-11-04T16:41:14Z

run benchmark tpcds

env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "9d3137dc3fcc4fbf7d0dccdbc506553a2ffb702d"

Rich-T-kid · 2026-11-04T16:41:16Z

run benchmark clickbench_partitioned

env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "9d3137dc3fcc4fbf7d0dccdbc506553a2ffb702d"

adriangbot · 2026-11-04T16:43:45Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6636310409-4466-z4n6x 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing 9d3137d (9d3137d) to ea33cce diff

Run configuration

run benchmark tpch
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "9d3137dc3fcc4fbf7d0dccdbc506553a2ffb702d"

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-11-04T16:44:44Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6636311892-4468-559d8 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing 9d3137d (9d3137d) to ea33cce diff

Run configuration

run benchmark clickbench_partitioned
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "9d3137dc3fcc4fbf7d0dccdbc506553a2ffb702d"

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-11-04T16:44:52Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6636311204-4469-n2vfd 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing 9d3137d (9d3137d) to ea33cce diff

Run configuration

run benchmark tpcds
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "9d3137dc3fcc4fbf7d0dccdbc506553a2ffb702d"

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-11-04T16:44:53Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6636309650-4467-jqk5q 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing 9d3137d (9d3137d) to ea33cce diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "9d3137dc3fcc4fbf7d0dccdbc506553a2ffb702d"

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-11-04T16:50:37Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing 9d3137d (9d3137d) to ea33cce diff

Run configuration

run benchmark tpch
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "9d3137dc3fcc4fbf7d0dccdbc506553a2ffb702d"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and rich-T-kid_dictionary-aggregate-emission
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃      HEAD ┃ rich-T-kid_dictionary-aggregate-emission ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │  40.31 ms │                                 39.68 ms │ no change │
│ QQuery 2  │  19.27 ms │                                 19.08 ms │ no change │
│ QQuery 3  │  28.89 ms │                                 28.97 ms │ no change │
│ QQuery 4  │  17.59 ms │                                 17.27 ms │ no change │
│ QQuery 5  │  36.27 ms │                                 36.11 ms │ no change │
│ QQuery 6  │  14.16 ms │                                 13.91 ms │ no change │
│ QQuery 7  │  42.26 ms │                                 43.17 ms │ no change │
│ QQuery 8  │  41.06 ms │                                 41.00 ms │ no change │
│ QQuery 9  │  51.15 ms │                                 50.24 ms │ no change │
│ QQuery 10 │  60.40 ms │                                 59.68 ms │ no change │
│ QQuery 11 │  14.07 ms │                                 14.05 ms │ no change │
│ QQuery 12 │  20.70 ms │                                 20.58 ms │ no change │
│ QQuery 13 │  47.58 ms │                                 47.86 ms │ no change │
│ QQuery 14 │  24.04 ms │                                 23.89 ms │ no change │
│ QQuery 15 │  31.48 ms │                                 31.15 ms │ no change │
│ QQuery 16 │  13.63 ms │                                 13.43 ms │ no change │
│ QQuery 17 │  71.64 ms │                                 71.06 ms │ no change │
│ QQuery 18 │ 115.76 ms │                                115.64 ms │ no change │
│ QQuery 19 │  29.20 ms │                                 29.03 ms │ no change │
│ QQuery 20 │  34.16 ms │                                 34.45 ms │ no change │
│ QQuery 21 │  54.51 ms │                                 54.30 ms │ no change │
│ QQuery 22 │  16.04 ms │                                 16.06 ms │ no change │
└───────────┴───────────┴──────────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                       ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                       │ 824.16ms │
│ Total Time (rich-T-kid_dictionary-aggregate-emission)   │ 820.59ms │
│ Average Time (HEAD)                                     │  37.46ms │
│ Average Time (rich-T-kid_dictionary-aggregate-emission) │  37.30ms │
│ Queries Faster                                          │        0 │
│ Queries Slower                                          │        0 │
│ Queries with No Change                                  │       22 │
│ Queries with Failure                                    │        0 │
└─────────────────────────────────────────────────────────┴──────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-T-kid_dictionary-aggregate-emission
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query     ┃                              HEAD ┃ rich-T-kid_dictionary-aggregate-emission ┃       Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 1  │    40.31 / 41.38 ±0.91 / 42.60 ms │           39.68 / 40.47 ±1.09 / 42.58 ms │    no change │
│ QQuery 2  │    19.27 / 19.48 ±0.18 / 19.77 ms │           19.08 / 19.32 ±0.25 / 19.71 ms │    no change │
│ QQuery 3  │    28.89 / 29.21 ±0.20 / 29.50 ms │           28.97 / 29.54 ±1.13 / 31.80 ms │    no change │
│ QQuery 4  │    17.59 / 18.01 ±0.59 / 19.18 ms │           17.27 / 17.48 ±0.13 / 17.66 ms │    no change │
│ QQuery 5  │    36.27 / 36.68 ±0.29 / 37.16 ms │           36.11 / 36.36 ±0.26 / 36.82 ms │    no change │
│ QQuery 6  │    14.16 / 14.56 ±0.33 / 15.04 ms │           13.91 / 14.04 ±0.10 / 14.19 ms │    no change │
│ QQuery 7  │    42.26 / 43.96 ±0.89 / 44.81 ms │           43.17 / 44.35 ±1.01 / 46.01 ms │    no change │
│ QQuery 8  │    41.06 / 41.98 ±1.19 / 44.29 ms │           41.00 / 42.14 ±1.12 / 43.88 ms │    no change │
│ QQuery 9  │    51.15 / 51.72 ±0.39 / 52.34 ms │           50.24 / 51.61 ±0.70 / 52.20 ms │    no change │
│ QQuery 10 │    60.40 / 61.26 ±0.91 / 62.88 ms │           59.68 / 60.86 ±0.96 / 62.40 ms │    no change │
│ QQuery 11 │    14.07 / 14.39 ±0.25 / 14.79 ms │           14.05 / 15.16 ±1.26 / 17.12 ms │ 1.05x slower │
│ QQuery 12 │    20.70 / 21.03 ±0.26 / 21.32 ms │           20.58 / 20.89 ±0.16 / 21.00 ms │    no change │
│ QQuery 13 │    47.58 / 49.69 ±2.66 / 54.94 ms │           47.86 / 49.00 ±1.39 / 51.64 ms │    no change │
│ QQuery 14 │    24.04 / 24.34 ±0.24 / 24.79 ms │           23.89 / 24.12 ±0.15 / 24.29 ms │    no change │
│ QQuery 15 │    31.48 / 31.65 ±0.15 / 31.91 ms │           31.15 / 31.91 ±0.62 / 32.78 ms │    no change │
│ QQuery 16 │    13.63 / 13.74 ±0.08 / 13.84 ms │           13.43 / 13.58 ±0.10 / 13.70 ms │    no change │
│ QQuery 17 │    71.64 / 72.27 ±0.81 / 73.73 ms │           71.06 / 72.15 ±1.03 / 74.10 ms │    no change │
│ QQuery 18 │ 115.76 / 118.62 ±1.94 / 121.64 ms │        115.64 / 119.13 ±2.30 / 121.51 ms │    no change │
│ QQuery 19 │    29.20 / 31.28 ±2.29 / 35.56 ms │           29.03 / 29.84 ±1.12 / 32.05 ms │    no change │
│ QQuery 20 │    34.16 / 34.84 ±0.73 / 36.25 ms │           34.45 / 34.85 ±0.24 / 35.10 ms │    no change │
│ QQuery 21 │    54.51 / 57.06 ±3.37 / 63.70 ms │           54.30 / 56.94 ±2.07 / 60.46 ms │    no change │
│ QQuery 22 │    16.04 / 16.39 ±0.52 / 17.41 ms │           16.06 / 16.34 ±0.34 / 17.02 ms │    no change │
└───────────┴───────────────────────────────────┴──────────────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                                       ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                                       │ 843.54ms │
│ Total Time (rich-T-kid_dictionary-aggregate-emission)   │ 840.08ms │
│ Average Time (HEAD)                                     │  38.34ms │
│ Average Time (rich-T-kid_dictionary-aggregate-emission) │  38.19ms │
│ Queries Faster                                          │        0 │
│ Queries Slower                                          │        1 │
│ Queries with No Change                                  │       21 │
│ Queries with Failure                                    │        0 │
└─────────────────────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric	Value
Wall time	5.0s
Peak memory	1.4 GiB
Avg memory	718.3 MiB
CPU user	21.5s
CPU sys	1.6s
Peak spill	0 B

tpch — branch

Metric	Value
Wall time	5.0s
Peak memory	2.1 GiB
Avg memory	851.5 MiB
CPU user	21.6s
CPU sys	1.5s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-11-04T16:52:37Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing 9d3137d (9d3137d) to ea33cce diff

Run configuration

run benchmark tpcds
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "9d3137dc3fcc4fbf7d0dccdbc506553a2ffb702d"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and rich-T-kid_dictionary-aggregate-emission
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ rich-T-kid_dictionary-aggregate-emission ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │    5.78 ms │                                  5.60 ms │     no change │
│ QQuery 2  │   77.32 ms │                                 78.11 ms │     no change │
│ QQuery 3  │   29.45 ms │                                 29.42 ms │     no change │
│ QQuery 4  │  898.70 ms │                                911.53 ms │     no change │
│ QQuery 5  │   51.60 ms │                                 50.78 ms │     no change │
│ QQuery 6  │   34.63 ms │                                 33.61 ms │     no change │
│ QQuery 7  │   71.73 ms │                                 70.61 ms │     no change │
│ QQuery 8  │   39.40 ms │                                 38.78 ms │     no change │
│ QQuery 9  │   50.26 ms │                                 52.28 ms │     no change │
│ QQuery 10 │   59.04 ms │                                 59.24 ms │     no change │
│ QQuery 11 │  514.25 ms │                                523.95 ms │     no change │
│ QQuery 12 │   26.73 ms │                                 26.28 ms │     no change │
│ QQuery 13 │  112.89 ms │                                113.89 ms │     no change │
│ QQuery 14 │  382.38 ms │                                377.89 ms │     no change │
│ QQuery 15 │   53.81 ms │                                 55.70 ms │     no change │
│ QQuery 16 │    6.55 ms │                                  6.68 ms │     no change │
│ QQuery 17 │   78.65 ms │                                 78.81 ms │     no change │
│ QQuery 18 │  109.27 ms │                                106.04 ms │     no change │
│ QQuery 19 │   42.88 ms │                                 42.22 ms │     no change │
│ QQuery 20 │   31.74 ms │                                 31.97 ms │     no change │
│ QQuery 21 │   17.94 ms │                                 17.69 ms │     no change │
│ QQuery 22 │   75.25 ms │                                 73.23 ms │     no change │
│ QQuery 23 │  310.69 ms │                                303.58 ms │     no change │
│ QQuery 24 │  202.28 ms │                                203.54 ms │     no change │
│ QQuery 25 │  106.00 ms │                                107.69 ms │     no change │
│ QQuery 26 │   45.63 ms │                                 45.53 ms │     no change │
│ QQuery 27 │    6.77 ms │                                  6.70 ms │     no change │
│ QQuery 28 │   60.99 ms │                                 60.02 ms │     no change │
│ QQuery 29 │   94.43 ms │                                 97.74 ms │     no change │
│ QQuery 30 │   37.39 ms │                                 36.79 ms │     no change │
│ QQuery 31 │  113.12 ms │                                111.48 ms │     no change │
│ QQuery 32 │   20.88 ms │                                 21.31 ms │     no change │
│ QQuery 33 │   38.75 ms │                                 39.31 ms │     no change │
│ QQuery 34 │    9.89 ms │                                  9.85 ms │     no change │
│ QQuery 35 │   73.74 ms │                                 73.10 ms │     no change │
│ QQuery 36 │    6.27 ms │                                  6.25 ms │     no change │
│ QQuery 37 │    7.93 ms │                                  7.84 ms │     no change │
│ QQuery 38 │   67.15 ms │                                 66.64 ms │     no change │
│ QQuery 39 │   73.84 ms │                                 75.63 ms │     no change │
│ QQuery 40 │   23.79 ms │                                 23.94 ms │     no change │
│ QQuery 41 │   14.55 ms │                                 14.14 ms │     no change │
│ QQuery 42 │   24.22 ms │                                 24.28 ms │     no change │
│ QQuery 43 │    5.29 ms │                                  5.21 ms │     no change │
│ QQuery 44 │    8.85 ms │                                  9.18 ms │     no change │
│ QQuery 45 │   35.93 ms │                                 39.22 ms │  1.09x slower │
│ QQuery 46 │   11.66 ms │                                 11.60 ms │     no change │
│ QQuery 47 │  208.41 ms │                                207.10 ms │     no change │
│ QQuery 48 │   91.57 ms │                                 90.93 ms │     no change │
│ QQuery 49 │   71.33 ms │                                 70.80 ms │     no change │
│ QQuery 50 │   58.03 ms │                                 57.12 ms │     no change │
│ QQuery 51 │   94.64 ms │                                 95.70 ms │     no change │
│ QQuery 52 │   24.59 ms │                                 24.77 ms │     no change │
│ QQuery 53 │   29.48 ms │                                 29.79 ms │     no change │
│ QQuery 54 │   54.57 ms │                                 54.46 ms │     no change │
│ QQuery 55 │   24.20 ms │                                 24.08 ms │     no change │
│ QQuery 56 │   39.29 ms │                                 39.20 ms │     no change │
│ QQuery 57 │  153.85 ms │                                156.02 ms │     no change │
│ QQuery 58 │  106.55 ms │                                106.08 ms │     no change │
│ QQuery 59 │  108.00 ms │                                109.61 ms │     no change │
│ QQuery 60 │   43.91 ms │                                 43.90 ms │     no change │
│ QQuery 61 │   10.14 ms │                                  9.86 ms │     no change │
│ QQuery 62 │   39.07 ms │                                 38.61 ms │     no change │
│ QQuery 63 │   29.93 ms │                                 29.65 ms │     no change │
│ QQuery 64 │  365.85 ms │                                363.73 ms │     no change │
│ QQuery 65 │   79.46 ms │                                 80.99 ms │     no change │
│ QQuery 66 │   73.40 ms │                                 73.77 ms │     no change │
│ QQuery 67 │  243.64 ms │                                238.02 ms │     no change │
│ QQuery 68 │   11.93 ms │                                 12.25 ms │     no change │
│ QQuery 69 │   55.31 ms │                                 55.47 ms │     no change │
│ QQuery 70 │   99.50 ms │                                 99.81 ms │     no change │
│ QQuery 71 │   35.89 ms │                                 36.15 ms │     no change │
│ QQuery 72 │ 1721.11 ms │                               1613.71 ms │ +1.07x faster │
│ QQuery 73 │    9.80 ms │                                  9.85 ms │     no change │
│ QQuery 74 │  265.67 ms │                                280.19 ms │  1.05x slower │
│ QQuery 75 │  141.63 ms │                                141.60 ms │     no change │
│ QQuery 76 │   35.42 ms │                                 35.36 ms │     no change │
│ QQuery 77 │   60.56 ms │                                 60.49 ms │     no change │
│ QQuery 78 │  166.07 ms │                                164.93 ms │     no change │
│ QQuery 79 │   68.30 ms │                                 67.94 ms │     no change │
│ QQuery 80 │   97.55 ms │                                 95.60 ms │     no change │
│ QQuery 81 │   29.23 ms │                                 29.36 ms │     no change │
│ QQuery 82 │   17.83 ms │                                 18.01 ms │     no change │
│ QQuery 83 │   33.82 ms │                                 33.40 ms │     no change │
│ QQuery 84 │   31.38 ms │                                 31.40 ms │     no change │
│ QQuery 85 │   95.83 ms │                                 96.11 ms │     no change │
│ QQuery 86 │   25.16 ms │                                 25.64 ms │     no change │
│ QQuery 87 │   67.17 ms │                                 66.00 ms │     no change │
│ QQuery 88 │   60.58 ms │                                 59.46 ms │     no change │
│ QQuery 89 │   36.85 ms │                                 36.18 ms │     no change │
│ QQuery 90 │   16.90 ms │                                 16.40 ms │     no change │
│ QQuery 91 │   40.26 ms │                                 39.62 ms │     no change │
│ QQuery 92 │   30.11 ms │                                 30.19 ms │     no change │
│ QQuery 93 │   51.74 ms │                                 51.03 ms │     no change │
│ QQuery 94 │   40.47 ms │                                 39.12 ms │     no change │
│ QQuery 95 │   82.48 ms │                                 81.25 ms │     no change │
│ QQuery 96 │   24.35 ms │                                 23.97 ms │     no change │
│ QQuery 97 │   52.91 ms │                                 51.37 ms │     no change │
│ QQuery 98 │   39.00 ms │                                 39.49 ms │     no change │
│ QQuery 99 │   58.07 ms │                                 58.25 ms │     no change │
└───────────┴────────────┴──────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                       ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                       │ 9723.07ms │
│ Total Time (rich-T-kid_dictionary-aggregate-emission)   │ 9628.65ms │
│ Average Time (HEAD)                                     │   98.21ms │
│ Average Time (rich-T-kid_dictionary-aggregate-emission) │   97.26ms │
│ Queries Faster                                          │         1 │
│ Queries Slower                                          │         2 │
│ Queries with No Change                                  │        96 │
│ Queries with Failure                                    │         0 │
└─────────────────────────────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-T-kid_dictionary-aggregate-emission
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃ rich-T-kid_dictionary-aggregate-emission ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.78 / 6.34 ±0.96 / 8.25 ms │              5.60 / 6.17 ±0.92 / 8.00 ms │     no change │
│ QQuery 2  │        77.32 / 77.99 ±0.34 / 78.26 ms │           78.11 / 78.28 ±0.15 / 78.50 ms │     no change │
│ QQuery 3  │        29.45 / 29.57 ±0.15 / 29.84 ms │           29.42 / 29.60 ±0.11 / 29.75 ms │     no change │
│ QQuery 4  │    898.70 / 921.69 ±13.00 / 937.68 ms │        911.53 / 917.95 ±6.84 / 928.21 ms │     no change │
│ QQuery 5  │        51.60 / 52.13 ±0.76 / 53.65 ms │           50.78 / 51.37 ±0.42 / 52.06 ms │     no change │
│ QQuery 6  │        34.63 / 34.99 ±0.26 / 35.42 ms │           33.61 / 34.12 ±0.55 / 35.18 ms │     no change │
│ QQuery 7  │        71.73 / 72.27 ±0.66 / 73.55 ms │           70.61 / 71.43 ±0.47 / 72.00 ms │     no change │
│ QQuery 8  │        39.40 / 40.90 ±2.40 / 45.69 ms │           38.78 / 39.81 ±0.58 / 40.34 ms │     no change │
│ QQuery 9  │        50.26 / 53.57 ±2.70 / 57.57 ms │           52.28 / 54.97 ±2.35 / 59.30 ms │     no change │
│ QQuery 10 │        59.04 / 59.82 ±0.52 / 60.67 ms │           59.24 / 59.55 ±0.28 / 60.02 ms │     no change │
│ QQuery 11 │    514.25 / 527.55 ±11.23 / 543.75 ms │        523.95 / 530.38 ±3.44 / 533.93 ms │     no change │
│ QQuery 12 │        26.73 / 26.81 ±0.08 / 26.98 ms │           26.28 / 26.75 ±0.41 / 27.39 ms │     no change │
│ QQuery 13 │     112.89 / 114.06 ±0.71 / 115.05 ms │        113.89 / 114.49 ±0.52 / 115.39 ms │     no change │
│ QQuery 14 │     382.38 / 384.66 ±2.45 / 389.34 ms │        377.89 / 386.79 ±6.12 / 395.70 ms │     no change │
│ QQuery 15 │        53.81 / 55.14 ±0.94 / 56.68 ms │           55.70 / 57.20 ±1.90 / 60.78 ms │     no change │
│ QQuery 16 │           6.55 / 6.61 ±0.07 / 6.73 ms │              6.68 / 6.71 ±0.02 / 6.73 ms │     no change │
│ QQuery 17 │        78.65 / 80.02 ±1.19 / 82.20 ms │           78.81 / 80.44 ±1.17 / 82.10 ms │     no change │
│ QQuery 18 │     109.27 / 109.95 ±0.48 / 110.78 ms │        106.04 / 109.79 ±2.87 / 114.32 ms │     no change │
│ QQuery 19 │        42.88 / 43.21 ±0.30 / 43.76 ms │           42.22 / 42.62 ±0.29 / 43.00 ms │     no change │
│ QQuery 20 │        31.74 / 32.37 ±0.45 / 33.02 ms │           31.97 / 32.44 ±0.32 / 32.92 ms │     no change │
│ QQuery 21 │        17.94 / 18.01 ±0.05 / 18.08 ms │           17.69 / 17.86 ±0.16 / 18.15 ms │     no change │
│ QQuery 22 │        75.25 / 76.48 ±1.00 / 78.31 ms │           73.23 / 74.96 ±1.39 / 76.40 ms │     no change │
│ QQuery 23 │     310.69 / 312.17 ±2.00 / 316.09 ms │        303.58 / 312.83 ±6.16 / 318.95 ms │     no change │
│ QQuery 24 │     202.28 / 206.39 ±4.95 / 215.09 ms │        203.54 / 207.46 ±2.66 / 211.91 ms │     no change │
│ QQuery 25 │     106.00 / 108.14 ±2.48 / 112.95 ms │        107.69 / 109.31 ±1.65 / 111.84 ms │     no change │
│ QQuery 26 │        45.63 / 45.94 ±0.18 / 46.20 ms │           45.53 / 45.80 ±0.21 / 46.14 ms │     no change │
│ QQuery 27 │           6.77 / 6.85 ±0.05 / 6.92 ms │              6.70 / 6.73 ±0.02 / 6.76 ms │     no change │
│ QQuery 28 │        60.99 / 61.69 ±0.57 / 62.41 ms │           60.02 / 62.00 ±1.05 / 63.04 ms │     no change │
│ QQuery 29 │       94.43 / 96.88 ±2.29 / 100.14 ms │          97.74 / 99.58 ±1.36 / 101.53 ms │     no change │
│ QQuery 30 │        37.39 / 37.70 ±0.17 / 37.91 ms │           36.79 / 37.38 ±0.36 / 37.87 ms │     no change │
│ QQuery 31 │     113.12 / 114.95 ±2.36 / 119.51 ms │        111.48 / 113.96 ±4.24 / 122.41 ms │     no change │
│ QQuery 32 │        20.88 / 21.35 ±0.39 / 22.00 ms │           21.31 / 22.13 ±1.26 / 24.61 ms │     no change │
│ QQuery 33 │        38.75 / 39.25 ±0.30 / 39.66 ms │           39.31 / 40.00 ±0.76 / 41.48 ms │     no change │
│ QQuery 34 │         9.89 / 10.12 ±0.16 / 10.36 ms │             9.85 / 9.98 ±0.15 / 10.24 ms │     no change │
│ QQuery 35 │        73.74 / 74.69 ±0.75 / 75.84 ms │           73.10 / 74.37 ±0.83 / 75.31 ms │     no change │
│ QQuery 36 │           6.27 / 6.38 ±0.08 / 6.52 ms │              6.25 / 6.36 ±0.11 / 6.57 ms │     no change │
│ QQuery 37 │           7.93 / 7.98 ±0.05 / 8.06 ms │              7.84 / 7.95 ±0.15 / 8.23 ms │     no change │
│ QQuery 38 │        67.15 / 67.84 ±0.85 / 69.50 ms │           66.64 / 66.90 ±0.20 / 67.18 ms │     no change │
│ QQuery 39 │        73.84 / 74.68 ±0.53 / 75.40 ms │           75.63 / 76.89 ±1.02 / 78.38 ms │     no change │
│ QQuery 40 │        23.79 / 24.37 ±0.44 / 25.15 ms │           23.94 / 24.50 ±0.38 / 24.91 ms │     no change │
│ QQuery 41 │        14.55 / 14.71 ±0.09 / 14.82 ms │           14.14 / 14.27 ±0.14 / 14.48 ms │     no change │
│ QQuery 42 │        24.22 / 24.54 ±0.31 / 25.13 ms │           24.28 / 25.15 ±0.78 / 26.32 ms │     no change │
│ QQuery 43 │           5.29 / 5.43 ±0.09 / 5.54 ms │              5.21 / 5.35 ±0.09 / 5.43 ms │     no change │
│ QQuery 44 │           8.85 / 9.14 ±0.22 / 9.49 ms │            9.18 / 10.69 ±2.85 / 16.39 ms │  1.17x slower │
│ QQuery 45 │        35.93 / 38.09 ±1.76 / 40.95 ms │           39.22 / 40.22 ±0.92 / 41.93 ms │  1.06x slower │
│ QQuery 46 │        11.66 / 11.77 ±0.09 / 11.92 ms │           11.60 / 11.79 ±0.15 / 12.06 ms │     no change │
│ QQuery 47 │     208.41 / 216.52 ±6.13 / 224.54 ms │        207.10 / 214.62 ±4.84 / 221.53 ms │     no change │
│ QQuery 48 │        91.57 / 93.19 ±1.72 / 96.52 ms │           90.93 / 92.40 ±1.01 / 93.73 ms │     no change │
│ QQuery 49 │        71.33 / 72.69 ±1.28 / 74.96 ms │           70.80 / 73.15 ±2.15 / 76.95 ms │     no change │
│ QQuery 50 │        58.03 / 58.49 ±0.32 / 59.04 ms │           57.12 / 58.42 ±0.94 / 59.85 ms │     no change │
│ QQuery 51 │       94.64 / 97.31 ±2.41 / 101.82 ms │           95.70 / 97.71 ±1.51 / 99.53 ms │     no change │
│ QQuery 52 │        24.59 / 26.50 ±3.49 / 33.49 ms │           24.77 / 26.86 ±3.44 / 33.67 ms │     no change │
│ QQuery 53 │        29.48 / 29.85 ±0.25 / 30.21 ms │           29.79 / 30.07 ±0.23 / 30.46 ms │     no change │
│ QQuery 54 │        54.57 / 55.06 ±0.53 / 56.06 ms │           54.46 / 55.01 ±0.57 / 56.11 ms │     no change │
│ QQuery 55 │        24.20 / 24.50 ±0.19 / 24.81 ms │           24.08 / 24.26 ±0.22 / 24.69 ms │     no change │
│ QQuery 56 │        39.29 / 39.75 ±0.43 / 40.33 ms │           39.20 / 39.91 ±0.59 / 40.87 ms │     no change │
│ QQuery 57 │     153.85 / 155.81 ±2.73 / 161.11 ms │        156.02 / 159.90 ±3.58 / 166.34 ms │     no change │
│ QQuery 58 │     106.55 / 108.71 ±2.99 / 114.60 ms │        106.08 / 107.71 ±1.56 / 110.54 ms │     no change │
│ QQuery 59 │     108.00 / 108.57 ±0.55 / 109.44 ms │        109.61 / 109.96 ±0.32 / 110.52 ms │     no change │
│ QQuery 60 │        43.91 / 44.37 ±0.25 / 44.62 ms │           43.90 / 44.29 ±0.38 / 45.00 ms │     no change │
│ QQuery 61 │        10.14 / 10.25 ±0.09 / 10.40 ms │             9.86 / 9.96 ±0.10 / 10.12 ms │     no change │
│ QQuery 62 │        39.07 / 41.40 ±2.75 / 46.49 ms │           38.61 / 40.85 ±2.03 / 43.86 ms │     no change │
│ QQuery 63 │        29.93 / 30.43 ±0.40 / 31.08 ms │           29.65 / 29.92 ±0.24 / 30.34 ms │     no change │
│ QQuery 64 │     365.85 / 374.54 ±6.77 / 384.53 ms │        363.73 / 370.71 ±4.91 / 376.55 ms │     no change │
│ QQuery 65 │        79.46 / 81.22 ±1.51 / 82.95 ms │           80.99 / 81.90 ±0.90 / 83.42 ms │     no change │
│ QQuery 66 │        73.40 / 74.28 ±0.70 / 75.36 ms │           73.77 / 74.13 ±0.33 / 74.55 ms │     no change │
│ QQuery 67 │     243.64 / 249.59 ±3.87 / 255.80 ms │        238.02 / 241.94 ±3.85 / 249.33 ms │     no change │
│ QQuery 68 │        11.93 / 12.01 ±0.07 / 12.11 ms │           12.25 / 12.37 ±0.10 / 12.49 ms │     no change │
│ QQuery 69 │        55.31 / 56.01 ±1.21 / 58.43 ms │           55.47 / 55.74 ±0.23 / 56.13 ms │     no change │
│ QQuery 70 │      99.50 / 100.38 ±0.77 / 101.55 ms │         99.81 / 100.50 ±0.89 / 102.26 ms │     no change │
│ QQuery 71 │        35.89 / 38.97 ±5.57 / 50.09 ms │           36.15 / 37.13 ±0.87 / 38.52 ms │     no change │
│ QQuery 72 │ 1721.11 / 1779.47 ±48.46 / 1856.84 ms │    1613.71 / 1718.07 ±76.07 / 1837.27 ms │     no change │
│ QQuery 73 │         9.80 / 10.06 ±0.20 / 10.33 ms │             9.85 / 9.97 ±0.09 / 10.10 ms │     no change │
│ QQuery 74 │     265.67 / 274.80 ±6.44 / 283.31 ms │        280.19 / 286.49 ±3.58 / 290.09 ms │     no change │
│ QQuery 75 │     141.63 / 145.24 ±3.54 / 151.53 ms │        141.60 / 142.58 ±0.65 / 143.27 ms │     no change │
│ QQuery 76 │        35.42 / 35.68 ±0.16 / 35.85 ms │           35.36 / 38.34 ±3.59 / 45.34 ms │  1.07x slower │
│ QQuery 77 │        60.56 / 61.01 ±0.40 / 61.70 ms │           60.49 / 61.38 ±0.65 / 62.17 ms │     no change │
│ QQuery 78 │     166.07 / 175.34 ±6.14 / 182.64 ms │        164.93 / 170.12 ±4.89 / 179.04 ms │     no change │
│ QQuery 79 │        68.30 / 69.98 ±1.83 / 73.43 ms │           67.94 / 68.50 ±0.47 / 69.09 ms │     no change │
│ QQuery 80 │      97.55 / 100.22 ±3.20 / 106.46 ms │         95.60 / 100.48 ±6.73 / 113.42 ms │     no change │
│ QQuery 81 │        29.23 / 29.42 ±0.18 / 29.74 ms │           29.36 / 30.02 ±0.40 / 30.53 ms │     no change │
│ QQuery 82 │        17.83 / 18.57 ±0.88 / 20.28 ms │           18.01 / 18.20 ±0.19 / 18.49 ms │     no change │
│ QQuery 83 │        33.82 / 33.97 ±0.17 / 34.26 ms │           33.40 / 33.63 ±0.16 / 33.90 ms │     no change │
│ QQuery 84 │        31.38 / 31.80 ±0.24 / 32.04 ms │           31.40 / 31.76 ±0.31 / 32.13 ms │     no change │
│ QQuery 85 │       95.83 / 98.61 ±2.66 / 103.38 ms │          96.11 / 99.29 ±3.41 / 105.41 ms │     no change │
│ QQuery 86 │        25.16 / 25.45 ±0.27 / 25.96 ms │           25.64 / 26.05 ±0.59 / 27.23 ms │     no change │
│ QQuery 87 │        67.17 / 67.28 ±0.09 / 67.40 ms │           66.00 / 67.55 ±0.86 / 68.51 ms │     no change │
│ QQuery 88 │        60.58 / 63.70 ±3.63 / 70.53 ms │           59.46 / 60.07 ±0.33 / 60.39 ms │ +1.06x faster │
│ QQuery 89 │        36.85 / 37.16 ±0.28 / 37.65 ms │           36.18 / 37.89 ±2.56 / 42.86 ms │     no change │
│ QQuery 90 │        16.90 / 17.06 ±0.15 / 17.32 ms │           16.40 / 16.56 ±0.15 / 16.82 ms │     no change │
│ QQuery 91 │        40.26 / 40.65 ±0.31 / 41.13 ms │           39.62 / 39.89 ±0.18 / 40.09 ms │     no change │
│ QQuery 92 │        30.11 / 30.52 ±0.35 / 31.13 ms │           30.19 / 30.90 ±0.43 / 31.48 ms │     no change │
│ QQuery 93 │        51.74 / 52.93 ±2.03 / 56.97 ms │           51.03 / 52.03 ±0.72 / 53.02 ms │     no change │
│ QQuery 94 │        40.47 / 40.79 ±0.24 / 41.16 ms │           39.12 / 40.95 ±1.96 / 44.48 ms │     no change │
│ QQuery 95 │        82.48 / 83.52 ±0.88 / 85.05 ms │           81.25 / 83.56 ±2.30 / 87.47 ms │     no change │
│ QQuery 96 │        24.35 / 24.48 ±0.12 / 24.62 ms │           23.97 / 24.17 ±0.16 / 24.44 ms │     no change │
│ QQuery 97 │        52.91 / 54.91 ±2.12 / 58.61 ms │           51.37 / 52.09 ±1.26 / 54.60 ms │ +1.05x faster │
│ QQuery 98 │        39.00 / 39.81 ±0.50 / 40.45 ms │           39.49 / 39.59 ±0.11 / 39.81 ms │     no change │
│ QQuery 99 │        58.07 / 58.93 ±0.54 / 59.59 ms │           58.25 / 59.73 ±2.68 / 65.09 ms │     no change │
└───────────┴───────────────────────────────────────┴──────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                                       ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                                       │ 9946.96ms │
│ Total Time (rich-T-kid_dictionary-aggregate-emission)   │ 9884.54ms │
│ Average Time (HEAD)                                     │  100.47ms │
│ Average Time (rich-T-kid_dictionary-aggregate-emission) │   99.84ms │
│ Queries Faster                                          │         2 │
│ Queries Slower                                          │         3 │
│ Queries with No Change                                  │        94 │
│ Queries with Failure                                    │         0 │
└─────────────────────────────────────────────────────────┴───────────┘

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	55.0s
Peak memory	1.9 GiB
Avg memory	1.3 GiB
CPU user	200.0s
CPU sys	5.8s
Peak spill	0 B

tpcds — branch

Metric	Value
Wall time	50.0s
Peak memory	1.8 GiB
Avg memory	1.3 GiB
CPU user	197.3s
CPU sys	5.6s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-11-04T17:01:00Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing 9d3137d (9d3137d) to ea33cce diff

Run configuration

run benchmark dictionary_group_values
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "9d3137dc3fcc4fbf7d0dccdbc506553a2ffb702d"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

group                                                                             HEAD                                    rich-T-kid_dictionary-aggregate-emission
-----                                                                             ----                                    ----------------------------------------
dict_intern_emit/intern_emit/size_65536_card_1000_null_0.00                       1.01    488.8±7.63µs 127.9 MElem/sec    1.00    484.9±6.63µs 128.9 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_20_null_0.00                         1.02    424.2±7.76µs 147.3 MElem/sec    1.00    415.3±5.95µs 150.5 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_300_null_0.00                        1.01    452.5±8.17µs 138.1 MElem/sec    1.00    448.7±7.04µs 139.3 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_65536_null_0.00                      1.00      6.2±0.07ms 10.1 MElem/sec     1.02      6.3±0.07ms  9.9 MElem/sec
dict_intern_emit/intern_emit/size_65536_card_75_null_0.00                         1.00    439.6±6.27µs 142.2 MElem/sec    1.00    439.0±5.77µs 142.4 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_1000_null_0.00                        1.13  126.1±119.13µs 61.9 MElem/sec     1.00    112.0±1.46µs 69.8 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_20_null_0.00                          1.04     58.2±1.66µs 134.1 MElem/sec    1.00     55.8±1.43µs 140.0 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_300_null_0.00                         1.03     74.7±1.66µs 104.6 MElem/sec    1.00     72.5±1.73µs 107.7 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_75_null_0.00                          1.05     63.4±2.19µs 123.2 MElem/sec    1.00     60.4±1.22µs 129.4 MElem/sec
dict_intern_emit/intern_emit/size_8192_card_8192_null_0.00                        1.00    522.8±2.98µs 14.9 MElem/sec     1.00    524.3±2.20µs 14.9 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_1000_null_0.10     1.02      2.4±0.01ms 105.9 MElem/sec    1.00      2.3±0.01ms 107.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_20_null_0.10       1.00      2.2±0.01ms 115.3 MElem/sec    1.00      2.2±0.01ms 115.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_300_null_0.10      1.00      2.3±0.01ms 110.2 MElem/sec    1.00      2.3±0.01ms 110.5 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_65536_null_0.10    1.15     22.8±0.92ms 10.9 MElem/sec     1.00     19.9±0.75ms 12.5 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_65536_card_75_null_0.10       1.01      2.3±0.01ms 110.7 MElem/sec    1.00      2.2±0.01ms 111.8 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_1000_null_0.10      1.04    398.2±2.39µs 78.5 MElem/sec     1.00    383.9±1.51µs 81.4 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_20_null_0.10        1.01    248.9±2.36µs 125.6 MElem/sec    1.00    246.6±1.27µs 126.7 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_300_null_0.10       1.02    295.2±2.68µs 105.9 MElem/sec    1.00    289.2±1.77µs 108.1 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_75_null_0.10        1.02    265.3±2.85µs 117.8 MElem/sec    1.00    261.3±1.98µs 119.6 MElem/sec
dict_repeated_intern_emit/repeated_intern_emit/size_8192_card_8192_null_0.10      1.08  1310.0±10.64µs 23.9 MElem/sec     1.00   1217.7±3.88µs 25.7 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_1000_null_0.10       1.02   1359.3±2.62µs 46.0 MElem/sec     1.00   1339.1±3.99µs 46.7 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_20_null_0.10         1.05   1261.8±4.04µs 49.5 MElem/sec     1.00   1200.4±4.39µs 52.1 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_300_null_0.10        1.03   1263.6±3.54µs 49.5 MElem/sec     1.00   1230.2±2.93µs 50.8 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_65536_null_0.10      1.09      7.6±0.11ms  8.2 MElem/sec     1.00      6.9±0.07ms  9.0 MElem/sec
dict_scalar_append_equal/scalar_append_equal/size_65536_card_75_null_0.10         1.02   1247.2±5.14µs 50.1 MElem/sec     1.00   1219.7±3.00µs 51.2 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_100000                    1.09     23.2±1.24ms 10.8 MElem/sec     1.00     21.3±1.25ms 11.8 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_32768                     1.03      7.9±0.08ms 31.6 MElem/sec     1.00      7.7±0.07ms 32.5 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_500000                    1.02     81.1±4.06ms  3.1 MElem/sec     1.00     79.5±4.75ms  3.1 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_65536                     1.03     15.2±0.59ms 16.5 MElem/sec     1.00     14.7±0.49ms 17.0 MElem/sec
dict_shared_values_arc/shared_values_arc/size_8192_card_8192                      1.04      2.9±0.03ms 85.9 MElem/sec     1.00      2.8±0.02ms 88.9 MElem/sec
dict_take_n/take_n/size_65536_card_1000_null_0.10                                 1.00      2.5±0.01ms 101.5 MElem/sec    1.00      2.5±0.00ms 101.9 MElem/sec
dict_take_n/take_n/size_65536_card_20_null_0.10                                   1.00      2.1±0.01ms 121.1 MElem/sec    1.00      2.1±0.01ms 121.2 MElem/sec
dict_take_n/take_n/size_65536_card_300_null_0.10                                  1.00      2.2±0.01ms 111.8 MElem/sec    1.00      2.2±0.01ms 111.8 MElem/sec
dict_take_n/take_n/size_65536_card_65536_null_0.10                                1.00     26.7±0.34ms  9.4 MElem/sec     1.01     27.0±0.23ms  9.3 MElem/sec
dict_take_n/take_n/size_65536_card_75_null_0.10                                   1.00      2.2±0.01ms 115.0 MElem/sec    1.00      2.2±0.01ms 115.4 MElem/sec

Resource Usage

dictionary_group_values — base (merge-base)

Metric	Value
Wall time	440.1s
Peak memory	896.8 MiB
Avg memory	96.7 MiB
CPU user	415.0s
CPU sys	23.2s
Peak spill	0 B

dictionary_group_values — branch

Metric	Value
Wall time	515.1s
Peak memory	902.2 MiB
Avg memory	190.2 MiB
CPU user	429.0s
CPU sys	14.5s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-11-04T17:05:20Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing 9d3137d (9d3137d) to ea33cce diff

Run configuration

run benchmark clickbench_partitioned
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "ea33cce29c27ffb583e613d4a5900e9614be0fa8"
changed:
  ref: "9d3137dc3fcc4fbf7d0dccdbc506553a2ffb702d"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and rich-T-kid_dictionary-aggregate-emission
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query     ┃        HEAD ┃ rich-T-kid_dictionary-aggregate-emission ┃       Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 0  │     1.08 ms │                                  1.10 ms │    no change │
│ QQuery 1  │    11.21 ms │                                 11.20 ms │    no change │
│ QQuery 2  │    34.54 ms │                                 34.18 ms │    no change │
│ QQuery 3  │    30.69 ms │                                 30.25 ms │    no change │
│ QQuery 4  │   238.00 ms │                                236.37 ms │    no change │
│ QQuery 5  │  1433.78 ms │                               1467.96 ms │    no change │
│ QQuery 6  │     1.13 ms │                                  1.15 ms │    no change │
│ QQuery 7  │    12.38 ms │                                 12.34 ms │    no change │
│ QQuery 8  │   340.35 ms │                                340.45 ms │    no change │
│ QQuery 9  │   471.70 ms │                                480.17 ms │    no change │
│ QQuery 10 │   202.08 ms │                                199.07 ms │    no change │
│ QQuery 11 │   212.51 ms │                                209.59 ms │    no change │
│ QQuery 12 │  1504.83 ms │                               1510.15 ms │    no change │
│ QQuery 13 │  2036.97 ms │                               2070.21 ms │    no change │
│ QQuery 14 │  1630.41 ms │                               1665.02 ms │    no change │
│ QQuery 15 │   284.82 ms │                                292.61 ms │    no change │
│ QQuery 16 │  3707.89 ms │                               3692.69 ms │    no change │
│ QQuery 17 │  3663.66 ms │                               3730.62 ms │    no change │
│ QQuery 18 │  6763.83 ms │                               6442.15 ms │    no change │
│ QQuery 19 │    27.12 ms │                                 27.02 ms │    no change │
│ QQuery 20 │  1051.46 ms │                               1048.93 ms │    no change │
│ QQuery 21 │  1377.77 ms │                               1373.02 ms │    no change │
│ QQuery 22 │  3265.85 ms │                               3238.21 ms │    no change │
│ QQuery 23 │  9325.63 ms │                               9259.68 ms │    no change │
│ QQuery 24 │    95.28 ms │                                 95.61 ms │    no change │
│ QQuery 25 │   352.48 ms │                                351.35 ms │    no change │
│ QQuery 26 │    96.60 ms │                                 96.54 ms │    no change │
│ QQuery 27 │  1147.61 ms │                               1157.07 ms │    no change │
│ QQuery 28 │  2776.52 ms │                               2769.55 ms │    no change │
│ QQuery 29 │    37.82 ms │                                 37.34 ms │    no change │
│ QQuery 30 │   550.69 ms │                                544.21 ms │    no change │
│ QQuery 31 │   538.52 ms │                                543.01 ms │    no change │
│ QQuery 32 │  1042.61 ms │                               1011.72 ms │    no change │
│ QQuery 33 │ 13816.85 ms │                              13928.22 ms │    no change │
│ QQuery 34 │ 13848.64 ms │                              14550.53 ms │ 1.05x slower │
│ QQuery 35 │   299.15 ms │                                297.32 ms │    no change │
│ QQuery 36 │   168.20 ms │                                168.67 ms │    no change │
│ QQuery 37 │    97.71 ms │                                 96.50 ms │    no change │
│ QQuery 38 │    82.54 ms │                                 80.95 ms │    no change │
│ QQuery 39 │   396.28 ms │                                389.37 ms │    no change │
│ QQuery 40 │    14.16 ms │                                 14.11 ms │    no change │
│ QQuery 41 │    13.39 ms │                                 13.40 ms │    no change │
│ QQuery 42 │    13.35 ms │                                 13.00 ms │    no change │
└───────────┴─────────────┴──────────────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                       ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                       │ 73018.11ms │
│ Total Time (rich-T-kid_dictionary-aggregate-emission)   │ 73532.59ms │
│ Average Time (HEAD)                                     │  1698.10ms │
│ Average Time (rich-T-kid_dictionary-aggregate-emission) │  1710.06ms │
│ Queries Faster                                          │          0 │
│ Queries Slower                                          │          1 │
│ Queries with No Change                                  │         42 │
│ Queries with Failure                                    │          0 │
└─────────────────────────────────────────────────────────┴────────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-T-kid_dictionary-aggregate-emission
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                      HEAD ┃  rich-T-kid_dictionary-aggregate-emission ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │               1.08 / 1.71 ±1.14 / 3.98 ms │               1.10 / 1.74 ±1.17 / 4.08 ms │     no change │
│ QQuery 1  │            11.21 / 11.32 ±0.09 / 11.43 ms │            11.20 / 11.40 ±0.17 / 11.66 ms │     no change │
│ QQuery 2  │            34.54 / 34.87 ±0.25 / 35.17 ms │            34.18 / 34.84 ±0.73 / 36.24 ms │     no change │
│ QQuery 3  │            30.69 / 31.13 ±0.67 / 32.45 ms │            30.25 / 30.48 ±0.23 / 30.86 ms │     no change │
│ QQuery 4  │         238.00 / 242.85 ±2.91 / 247.03 ms │         236.37 / 241.08 ±4.15 / 246.69 ms │     no change │
│ QQuery 5  │     1433.78 / 1487.20 ±36.53 / 1546.86 ms │     1467.96 / 1504.96 ±26.71 / 1545.99 ms │     no change │
│ QQuery 6  │               1.13 / 1.29 ±0.24 / 1.75 ms │               1.15 / 1.30 ±0.25 / 1.79 ms │     no change │
│ QQuery 7  │            12.38 / 12.68 ±0.17 / 12.90 ms │            12.34 / 16.61 ±8.39 / 33.39 ms │  1.31x slower │
│ QQuery 8  │         340.35 / 348.21 ±6.70 / 358.47 ms │        340.45 / 355.40 ±13.91 / 378.71 ms │     no change │
│ QQuery 9  │         471.70 / 485.80 ±9.95 / 496.67 ms │         480.17 / 489.10 ±6.88 / 498.32 ms │     no change │
│ QQuery 10 │         202.08 / 203.30 ±1.02 / 204.49 ms │        199.07 / 206.16 ±12.84 / 231.81 ms │     no change │
│ QQuery 11 │         212.51 / 213.88 ±0.90 / 214.84 ms │         209.59 / 210.43 ±0.93 / 212.21 ms │     no change │
│ QQuery 12 │     1504.83 / 1530.60 ±25.39 / 1570.36 ms │     1510.15 / 1544.54 ±25.24 / 1579.59 ms │     no change │
│ QQuery 13 │    2036.97 / 2249.13 ±194.25 / 2478.95 ms │    2070.21 / 2365.05 ±320.38 / 2856.32 ms │  1.05x slower │
│ QQuery 14 │     1630.41 / 1689.69 ±38.69 / 1745.73 ms │     1665.02 / 1692.28 ±16.19 / 1713.80 ms │     no change │
│ QQuery 15 │         284.82 / 297.48 ±6.74 / 303.24 ms │         292.61 / 300.30 ±7.26 / 311.13 ms │     no change │
│ QQuery 16 │     3707.89 / 3779.08 ±57.74 / 3865.05 ms │     3692.69 / 3758.86 ±56.62 / 3823.56 ms │     no change │
│ QQuery 17 │     3663.66 / 3741.41 ±46.09 / 3795.54 ms │     3730.62 / 3798.96 ±70.94 / 3932.23 ms │     no change │
│ QQuery 18 │    6763.83 / 7026.21 ±178.95 / 7214.50 ms │    6442.15 / 6848.49 ±223.60 / 7039.59 ms │     no change │
│ QQuery 19 │            27.12 / 31.92 ±9.07 / 50.05 ms │            27.02 / 27.80 ±0.72 / 28.82 ms │ +1.15x faster │
│ QQuery 20 │      1051.46 / 1067.27 ±9.87 / 1078.70 ms │     1048.93 / 1068.80 ±12.39 / 1081.36 ms │     no change │
│ QQuery 21 │      1377.77 / 1381.82 ±2.80 / 1385.98 ms │      1373.02 / 1377.51 ±4.20 / 1385.05 ms │     no change │
│ QQuery 22 │     3265.85 / 3318.30 ±42.40 / 3365.76 ms │     3238.21 / 3267.21 ±21.53 / 3300.20 ms │     no change │
│ QQuery 23 │    9325.63 / 9468.89 ±126.80 / 9663.21 ms │    9259.68 / 9503.11 ±143.22 / 9692.63 ms │     no change │
│ QQuery 24 │         95.28 / 131.99 ±45.77 / 208.51 ms │           95.61 / 98.13 ±2.87 / 103.59 ms │ +1.35x faster │
│ QQuery 25 │        352.48 / 365.70 ±21.36 / 408.31 ms │        351.35 / 359.28 ±11.17 / 381.43 ms │     no change │
│ QQuery 26 │           96.60 / 98.45 ±1.24 / 100.40 ms │           96.54 / 98.14 ±1.75 / 101.48 ms │     no change │
│ QQuery 27 │      1147.61 / 1152.19 ±5.43 / 1162.80 ms │      1157.07 / 1160.85 ±3.17 / 1165.21 ms │     no change │
│ QQuery 28 │     2776.52 / 2833.78 ±35.03 / 2864.06 ms │     2769.55 / 2804.39 ±29.16 / 2849.15 ms │     no change │
│ QQuery 29 │          37.82 / 57.13 ±37.20 / 131.51 ms │          37.34 / 61.92 ±32.60 / 118.88 ms │  1.08x slower │
│ QQuery 30 │         550.69 / 561.73 ±6.98 / 571.96 ms │         544.21 / 551.40 ±6.16 / 562.12 ms │     no change │
│ QQuery 31 │        538.52 / 564.87 ±13.89 / 579.92 ms │        543.01 / 563.82 ±19.83 / 595.27 ms │     no change │
│ QQuery 32 │    1042.61 / 1139.73 ±110.68 / 1353.75 ms │     1011.72 / 1099.97 ±80.57 / 1236.73 ms │     no change │
│ QQuery 33 │ 13816.85 / 14395.59 ±519.05 / 15263.88 ms │ 13928.22 / 14683.15 ±532.74 / 15511.35 ms │     no change │
│ QQuery 34 │ 13848.64 / 14322.24 ±304.57 / 14683.91 ms │ 14550.53 / 14758.40 ±137.50 / 14973.90 ms │     no change │
│ QQuery 35 │        299.15 / 332.39 ±38.05 / 387.24 ms │       297.32 / 363.13 ±119.95 / 602.82 ms │  1.09x slower │
│ QQuery 36 │        168.20 / 187.03 ±15.11 / 214.33 ms │        168.67 / 194.80 ±29.53 / 252.22 ms │     no change │
│ QQuery 37 │          97.71 / 104.98 ±4.92 / 112.56 ms │          96.50 / 100.51 ±4.02 / 108.26 ms │     no change │
│ QQuery 38 │            82.54 / 85.00 ±2.35 / 88.83 ms │            80.95 / 84.12 ±2.69 / 87.62 ms │     no change │
│ QQuery 39 │         396.28 / 404.50 ±6.90 / 415.92 ms │        389.37 / 409.60 ±13.30 / 429.92 ms │     no change │
│ QQuery 40 │            14.16 / 14.67 ±0.27 / 14.90 ms │            14.11 / 14.62 ±0.33 / 14.93 ms │     no change │
│ QQuery 41 │            13.39 / 13.70 ±0.25 / 14.09 ms │            13.40 / 13.73 ±0.53 / 14.78 ms │     no change │
│ QQuery 42 │            13.35 / 14.12 ±1.43 / 16.97 ms │            13.00 / 13.17 ±0.10 / 13.30 ms │ +1.07x faster │
└───────────┴───────────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                       ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                       │ 75435.82ms │
│ Total Time (rich-T-kid_dictionary-aggregate-emission)   │ 76089.53ms │
│ Average Time (HEAD)                                     │  1754.32ms │
│ Average Time (rich-T-kid_dictionary-aggregate-emission) │  1769.52ms │
│ Queries Faster                                          │          3 │
│ Queries Slower                                          │          4 │
│ Queries with No Change                                  │         36 │
│ Queries with Failure                                    │          0 │
└─────────────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	380.1s
Peak memory	13.2 GiB
Avg memory	6.2 GiB
CPU user	3326.0s
CPU sys	204.9s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	385.1s
Peak memory	13.1 GiB
Avg memory	6.0 GiB
CPU user	3311.0s
CPU sys	196.9s
Peak spill	0 B

File an issue against this benchmark runner

Rich-T-kid · 2026-11-04T18:05:44Z

run benchmark clickbench_partitioned

env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "d0d542bfd3f87bf266e7141b432bce2200066ae6"
changed:
  ref: "9d3137dc3fcc4fbf7d0dccdbc506553a2ffb702d"

adriangbot · 2026-11-04T18:09:04Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6638197125-4477-8n69f 6.12.105+ #1 SMP Sat Sep 19 07:55:25 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing 9d3137d (9d3137d) to d0d542b diff

Run configuration

run benchmark clickbench_partitioned
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "d0d542bfd3f87bf266e7141b432bce2200066ae6"
changed:
  ref: "9d3137dc3fcc4fbf7d0dccdbc506553a2ffb702d"

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-11-04T18:30:03Z

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

Comparing 9d3137d (9d3137d) to d0d542b diff

Run configuration

run benchmark clickbench_partitioned
env:
  DATAFUSION_EXECUTION_PARQUET_ENABLE_RLE_TO_DICTIONARY: "true"
baseline:
  ref: "d0d542bfd3f87bf266e7141b432bce2200066ae6"
changed:
  ref: "9d3137dc3fcc4fbf7d0dccdbc506553a2ffb702d"

CPU Details (lscpu)

Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Details

Comparing HEAD and rich-T-kid_dictionary-aggregate-emission
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query     ┃        HEAD ┃ rich-T-kid_dictionary-aggregate-emission ┃       Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 0  │     1.13 ms │                                  1.12 ms │    no change │
│ QQuery 1  │    11.24 ms │                                 11.27 ms │    no change │
│ QQuery 2  │    34.21 ms │                                 34.05 ms │    no change │
│ QQuery 3  │    30.53 ms │                                 30.35 ms │    no change │
│ QQuery 4  │   236.83 ms │                                238.61 ms │    no change │
│ QQuery 5  │  1452.88 ms │                               1448.13 ms │    no change │
│ QQuery 6  │     1.15 ms │                                  1.15 ms │    no change │
│ QQuery 7  │    12.49 ms │                                 12.47 ms │    no change │
│ QQuery 8  │   338.46 ms │                                337.78 ms │    no change │
│ QQuery 9  │   471.22 ms │                                464.98 ms │    no change │
│ QQuery 10 │   201.26 ms │                                199.59 ms │    no change │
│ QQuery 11 │   212.44 ms │                                210.61 ms │    no change │
│ QQuery 12 │  1501.52 ms │                               1496.43 ms │    no change │
│ QQuery 13 │  2108.97 ms │                               2632.20 ms │ 1.25x slower │
│ QQuery 14 │  1635.56 ms │                               1631.50 ms │    no change │
│ QQuery 15 │   287.11 ms │                                294.42 ms │    no change │
│ QQuery 16 │  3653.19 ms │                               3660.23 ms │    no change │
│ QQuery 17 │  3685.97 ms │                               3691.10 ms │    no change │
│ QQuery 18 │  6717.22 ms │                               6741.88 ms │    no change │
│ QQuery 19 │    27.04 ms │                                 27.08 ms │    no change │
│ QQuery 20 │  1051.88 ms │                               1047.52 ms │    no change │
│ QQuery 21 │  1360.11 ms │                               1362.46 ms │    no change │
│ QQuery 22 │  3253.12 ms │                               3214.04 ms │    no change │
│ QQuery 23 │  9127.59 ms │                               9164.28 ms │    no change │
│ QQuery 24 │    96.65 ms │                                 96.15 ms │    no change │
│ QQuery 25 │   353.40 ms │                                352.01 ms │    no change │
│ QQuery 26 │    96.54 ms │                                 96.35 ms │    no change │
│ QQuery 27 │  1142.95 ms │                               1150.26 ms │    no change │
│ QQuery 28 │  2790.47 ms │                               2788.28 ms │    no change │
│ QQuery 29 │    37.17 ms │                                 37.36 ms │    no change │
│ QQuery 30 │   547.41 ms │                                547.08 ms │    no change │
│ QQuery 31 │   535.67 ms │                                544.52 ms │    no change │
│ QQuery 32 │  1030.09 ms │                               1043.13 ms │    no change │
│ QQuery 33 │ 13386.85 ms │                              14104.56 ms │ 1.05x slower │
│ QQuery 34 │ 13868.71 ms │                              13436.16 ms │    no change │
│ QQuery 35 │   300.95 ms │                                295.66 ms │    no change │
│ QQuery 36 │   173.44 ms │                                172.24 ms │    no change │
│ QQuery 37 │    96.29 ms │                                 96.97 ms │    no change │
│ QQuery 38 │    79.89 ms │                                 79.80 ms │    no change │
│ QQuery 39 │   385.67 ms │                                389.99 ms │    no change │
│ QQuery 40 │    14.22 ms │                                 14.23 ms │    no change │
│ QQuery 41 │    13.61 ms │                                 13.58 ms │    no change │
│ QQuery 42 │    13.26 ms │                                 13.13 ms │    no change │
└───────────┴─────────────┴──────────────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                       ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                       │ 72376.34ms │
│ Total Time (rich-T-kid_dictionary-aggregate-emission)   │ 73224.72ms │
│ Average Time (HEAD)                                     │  1683.17ms │
│ Average Time (rich-T-kid_dictionary-aggregate-emission) │  1702.90ms │
│ Queries Faster                                          │          0 │
│ Queries Slower                                          │          2 │
│ Queries with No Change                                  │         41 │
│ Queries with Failure                                    │          0 │
└─────────────────────────────────────────────────────────┴────────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and rich-T-kid_dictionary-aggregate-emission
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                      HEAD ┃  rich-T-kid_dictionary-aggregate-emission ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │               1.13 / 1.76 ±1.15 / 4.06 ms │               1.12 / 1.72 ±1.09 / 3.89 ms │     no change │
│ QQuery 1  │            11.24 / 11.44 ±0.26 / 11.95 ms │            11.27 / 11.61 ±0.56 / 12.73 ms │     no change │
│ QQuery 2  │            34.21 / 34.71 ±0.29 / 35.03 ms │            34.05 / 34.37 ±0.21 / 34.60 ms │     no change │
│ QQuery 3  │            30.53 / 31.19 ±0.64 / 32.33 ms │            30.35 / 30.82 ±0.31 / 31.31 ms │     no change │
│ QQuery 4  │         236.83 / 240.04 ±2.56 / 244.51 ms │         238.61 / 241.65 ±1.70 / 243.86 ms │     no change │
│ QQuery 5  │     1452.88 / 1493.40 ±64.40 / 1621.67 ms │     1448.13 / 1531.10 ±61.82 / 1612.22 ms │     no change │
│ QQuery 6  │               1.15 / 1.31 ±0.23 / 1.77 ms │               1.15 / 1.32 ±0.24 / 1.79 ms │     no change │
│ QQuery 7  │            12.49 / 14.83 ±4.40 / 23.63 ms │            12.47 / 12.62 ±0.08 / 12.72 ms │ +1.18x faster │
│ QQuery 8  │         338.46 / 340.78 ±1.52 / 343.23 ms │        337.78 / 346.93 ±10.06 / 365.16 ms │     no change │
│ QQuery 9  │         471.22 / 480.07 ±9.13 / 496.98 ms │        464.98 / 487.02 ±13.98 / 505.08 ms │     no change │
│ QQuery 10 │        201.26 / 206.83 ±10.48 / 227.79 ms │         199.59 / 206.86 ±9.34 / 223.60 ms │     no change │
│ QQuery 11 │         212.44 / 212.71 ±0.32 / 213.33 ms │         210.61 / 211.19 ±0.57 / 212.29 ms │     no change │
│ QQuery 12 │     1501.52 / 1549.20 ±52.54 / 1643.15 ms │     1496.43 / 1554.85 ±36.27 / 1608.31 ms │     no change │
│ QQuery 13 │    2108.97 / 2672.50 ±314.28 / 2982.20 ms │     2632.20 / 2766.18 ±90.91 / 2901.45 ms │     no change │
│ QQuery 14 │     1635.56 / 1668.14 ±30.65 / 1707.11 ms │     1631.50 / 1666.51 ±35.28 / 1724.04 ms │     no change │
│ QQuery 15 │        287.11 / 303.11 ±10.57 / 317.02 ms │         294.42 / 303.90 ±9.81 / 319.37 ms │     no change │
│ QQuery 16 │     3653.19 / 3746.62 ±52.58 / 3791.57 ms │     3660.23 / 3714.13 ±76.20 / 3862.51 ms │     no change │
│ QQuery 17 │     3685.97 / 3752.78 ±57.79 / 3860.51 ms │     3691.10 / 3796.37 ±58.13 / 3845.12 ms │     no change │
│ QQuery 18 │    6717.22 / 7135.30 ±222.86 / 7372.16 ms │    6741.88 / 6916.38 ±146.00 / 7133.80 ms │     no change │
│ QQuery 19 │          27.04 / 42.42 ±29.39 / 101.18 ms │           27.08 / 36.85 ±18.83 / 74.50 ms │ +1.15x faster │
│ QQuery 20 │     1051.88 / 1068.56 ±11.39 / 1082.62 ms │      1047.52 / 1057.08 ±4.98 / 1061.73 ms │     no change │
│ QQuery 21 │      1360.11 / 1374.34 ±9.07 / 1388.60 ms │      1362.46 / 1371.17 ±7.76 / 1382.47 ms │     no change │
│ QQuery 22 │     3253.12 / 3289.67 ±24.34 / 3327.72 ms │     3214.04 / 3259.39 ±26.98 / 3288.33 ms │     no change │
│ QQuery 23 │    9127.59 / 9397.25 ±200.95 / 9742.51 ms │    9164.28 / 9318.59 ±150.88 / 9526.29 ms │     no change │
│ QQuery 24 │         96.65 / 106.51 ±17.80 / 142.05 ms │         96.15 / 111.63 ±21.67 / 152.47 ms │     no change │
│ QQuery 25 │         353.40 / 355.97 ±2.18 / 359.49 ms │         352.01 / 354.21 ±1.91 / 357.48 ms │     no change │
│ QQuery 26 │            96.54 / 98.36 ±1.06 / 99.87 ms │           96.35 / 97.88 ±1.34 / 100.21 ms │     no change │
│ QQuery 27 │     1142.95 / 1153.46 ±12.71 / 1178.31 ms │     1150.26 / 1165.58 ±10.99 / 1178.15 ms │     no change │
│ QQuery 28 │     2790.47 / 2844.63 ±65.48 / 2962.08 ms │     2788.28 / 2825.49 ±23.39 / 2847.52 ms │     no change │
│ QQuery 29 │            37.17 / 39.85 ±4.80 / 49.44 ms │            37.36 / 39.55 ±3.16 / 45.64 ms │     no change │
│ QQuery 30 │         547.41 / 553.86 ±8.48 / 569.49 ms │        547.08 / 557.36 ±11.03 / 571.46 ms │     no change │
│ QQuery 31 │        535.67 / 554.34 ±12.83 / 566.74 ms │         544.52 / 553.00 ±7.99 / 566.04 ms │     no change │
│ QQuery 32 │     1030.09 / 1061.47 ±33.97 / 1106.11 ms │     1043.13 / 1106.49 ±35.41 / 1150.06 ms │     no change │
│ QQuery 33 │ 13386.85 / 13813.04 ±284.84 / 14090.96 ms │ 14104.56 / 14795.79 ±412.34 / 15350.39 ms │  1.07x slower │
│ QQuery 34 │ 13868.71 / 14284.60 ±340.82 / 14748.00 ms │ 13436.16 / 13924.71 ±314.44 / 14304.50 ms │     no change │
│ QQuery 35 │       300.95 / 367.55 ±118.68 / 604.78 ms │        295.66 / 345.79 ±86.35 / 518.04 ms │ +1.06x faster │
│ QQuery 36 │        173.44 / 189.97 ±19.09 / 225.27 ms │        172.24 / 183.64 ±10.91 / 201.29 ms │     no change │
│ QQuery 37 │          96.29 / 103.84 ±9.37 / 121.03 ms │          96.97 / 101.56 ±2.92 / 105.44 ms │     no change │
│ QQuery 38 │            79.89 / 81.58 ±1.67 / 84.77 ms │            79.80 / 84.29 ±2.84 / 87.61 ms │     no change │
│ QQuery 39 │        385.67 / 410.88 ±19.74 / 433.93 ms │        389.99 / 408.15 ±11.95 / 423.91 ms │     no change │
│ QQuery 40 │            14.22 / 14.59 ±0.37 / 15.29 ms │            14.23 / 14.96 ±0.49 / 15.49 ms │     no change │
│ QQuery 41 │            13.61 / 13.84 ±0.15 / 14.03 ms │            13.58 / 15.84 ±4.13 / 24.11 ms │  1.14x slower │
│ QQuery 42 │            13.26 / 13.59 ±0.40 / 14.37 ms │            13.13 / 13.21 ±0.13 / 13.47 ms │     no change │
└───────────┴───────────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                                       ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                                       │ 75130.89ms │
│ Total Time (rich-T-kid_dictionary-aggregate-emission)   │ 75577.74ms │
│ Average Time (HEAD)                                     │  1747.23ms │
│ Average Time (rich-T-kid_dictionary-aggregate-emission) │  1757.62ms │
│ Queries Faster                                          │          3 │
│ Queries Slower                                          │          2 │
│ Queries with No Change                                  │         38 │
│ Queries with Failure                                    │          0 │
└─────────────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	380.1s
Peak memory	13.3 GiB
Avg memory	6.1 GiB
CPU user	3295.3s
CPU sys	194.1s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	380.1s
Peak memory	13.1 GiB
Avg memory	6.1 GiB
CPU user	3310.6s
CPU sys	196.1s
Peak spill	0 B

File an issue against this benchmark runner

github-actions Bot added the physical-plan Changes to the physical-plan crate label Nov 3, 2026

Rich-T-kid force-pushed the rich-T-kid/dictionary-aggregate-emission branch from 939ae6c to 4653896 Compare November 3, 2026 15:03

Rich-T-kid changed the title ~~perf: avoid rebuilding dictionary group keys on emission~~ perf: avoid rebuilding dictionary keys during group emission Nov 3, 2026

Rich-T-kid force-pushed the rich-T-kid/dictionary-aggregate-emission branch from 4653896 to b282594 Compare November 3, 2026 15:56

Rich-T-kid force-pushed the rich-T-kid/dictionary-aggregate-emission branch from b282594 to 5ee33fb Compare November 3, 2026 18:23

perf: validate dictionary keys at slot creation

9d3137d

Every group key comes from an already validated inner slot. Check the key width when allocating that slot so later key writes stay infallible and out of the per-row append path.

Conversation

Rich-T-kid commented Nov 3, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Why

What changed

Results

Uh oh!

Rich-T-kid commented Nov 3, 2026

Uh oh!

adriangbot commented Nov 3, 2026

Uh oh!

adriangbot commented Nov 3, 2026

Uh oh!

Rich-T-kid commented Nov 3, 2026

Uh oh!

Rich-T-kid commented Nov 3, 2026

Uh oh!

Rich-T-kid commented Nov 3, 2026

Uh oh!

Rich-T-kid commented Nov 3, 2026

Uh oh!

adriangbot commented Nov 3, 2026

Uh oh!

adriangbot commented Nov 3, 2026

Uh oh!

adriangbot commented Nov 3, 2026

Uh oh!

adriangbot commented Nov 3, 2026

Uh oh!

adriangbot commented Nov 3, 2026

Uh oh!

adriangbot commented Nov 3, 2026

Uh oh!

adriangbot commented Nov 3, 2026

Uh oh!

Rich-T-kid commented Nov 3, 2026

Uh oh!

Rich-T-kid commented Nov 3, 2026

Uh oh!

Rich-T-kid commented Nov 3, 2026

Uh oh!

adriangbot commented Nov 3, 2026

Uh oh!

adriangbot commented Nov 3, 2026

Uh oh!

adriangbot commented Nov 3, 2026

Uh oh!

adriangbot commented Nov 3, 2026

Uh oh!

adriangbot commented Nov 3, 2026

Uh oh!

adriangbot commented Nov 3, 2026

Uh oh!

Rich-T-kid commented Nov 3, 2026

Uh oh!

Rich-T-kid commented Nov 3, 2026

Uh oh!

adriangbot commented Nov 4, 2026

Uh oh!

adriangbot commented Nov 4, 2026

Uh oh!

adriangbot commented Nov 4, 2026

Uh oh!

adriangbot commented Nov 4, 2026

Uh oh!

adriangbot commented Nov 4, 2026

Uh oh!

Rich-T-kid commented Nov 4, 2026

Uh oh!

Rich-T-kid commented Nov 4, 2026

Uh oh!

Rich-T-kid commented Nov 4, 2026

Uh oh!

adriangbot commented Nov 4, 2026

Uh oh!

adriangbot commented Nov 4, 2026

Uh oh!

adriangbot commented Nov 4, 2026

Uh oh!

adriangbot commented Nov 4, 2026

Rich-T-kid commented Nov 3, 2026 •

edited

Loading