aggregate blocked impl by rluvaton · Pull Request #24928 · apache/datafusion

rluvaton · 2026-09-03T16:55:44Z

Huge blocked impl that is backwards compatible
And just see the performance cost

Currently it only contain blocked impl for single group by and some aggregate expression

Which issue does this PR close?

Closes #.

Rationale for this change

What changes are included in this PR?

What is the testing strategy for this PR?

Are there any user-facing changes?

…so all usages will be migrated

…warded

…ockedGroupColumn`

…not `usize`

…e grouped_hash_stream code

# Conflicts: # datafusion/expr-common/src/groups_accumulator.rs

rluvaton · 2026-09-03T16:55:53Z

run benchmarks

adriangbot · 2026-09-03T16:58:42Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5529143890-2119-gtnrh 6.12.94+ #1 SMP Fri Jul 17 09:42:57 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 add-blocks-impl-from-scratch (d9d288b) to d2b626c (merge-base) diff

Run configuration

run benchmark clickbench_partitioned

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-09-03T16:58:51Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5529143890-2120-q2wqb 6.12.94+ #1 SMP Fri Jul 17 09:42:57 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 add-blocks-impl-from-scratch (d9d288b) to d2b626c (merge-base) diff

Run configuration

run benchmark tpcds

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-09-03T16:59:03Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5529143890-2121-jff8g 6.12.94+ #1 SMP Fri Jul 17 09:42:57 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 add-blocks-impl-from-scratch (d9d288b) to d2b626c (merge-base) diff

Run configuration

run benchmark tpch

Results will be posted here when complete

File an issue against this benchmark runner

github-actions · 2026-09-03T17:12:47Z

Thank you for opening this pull request!

Reviewer note: cargo-semver-checks reported the current version number is not SemVer-compatible with the changes in this pull request (compared against the base branch).

Details

     Cloning apache/main
    Building datafusion v55.0.0 (current)
       Built [  50.277s] (current)
     Parsing datafusion v55.0.0 (current)
      Parsed [   0.029s] (current)
    Building datafusion v55.0.0 (baseline)
       Built [  48.248s] (baseline)
     Parsing datafusion v55.0.0 (baseline)
      Parsed [   0.029s] (baseline)
    Checking datafusion v55.0.0 -> v55.0.0 (no change; assume patch)
     Checked [   0.811s] 223 checks: 223 pass, 31 skip
     Summary no semver update required
    Finished [ 101.481s] datafusion
    Building datafusion-common v55.0.0 (current)
       Built [  27.442s] (current)
     Parsing datafusion-common v55.0.0 (current)
      Parsed [   0.053s] (current)
    Building datafusion-common v55.0.0 (baseline)
       Built [  27.225s] (baseline)
     Parsing datafusion-common v55.0.0 (baseline)
      Parsed [   0.054s] (baseline)
    Checking datafusion-common v55.0.0 -> v55.0.0 (no change; assume patch)
     Checked [   0.870s] 223 checks: 223 pass, 31 skip
     Summary no semver update required
    Finished [  56.985s] datafusion-common
    Building datafusion-expr v55.0.0 (current)
       Built [  23.763s] (current)
     Parsing datafusion-expr v55.0.0 (current)
      Parsed [   0.066s] (current)
    Building datafusion-expr v55.0.0 (baseline)
       Built [  23.628s] (baseline)
     Parsing datafusion-expr v55.0.0 (baseline)
      Parsed [   0.065s] (baseline)
    Checking datafusion-expr v55.0.0 -> v55.0.0 (no change; assume patch)
     Checked [   1.667s] 223 checks: 223 pass, 31 skip
     Summary no semver update required
    Finished [  50.292s] datafusion-expr
    Building datafusion-expr-common v55.0.0 (current)
       Built [  16.503s] (current)
     Parsing datafusion-expr-common v55.0.0 (current)
      Parsed [   0.024s] (current)
    Building datafusion-expr-common v55.0.0 (baseline)
       Built [  16.057s] (baseline)
     Parsing datafusion-expr-common v55.0.0 (baseline)
      Parsed [   0.016s] (baseline)
    Checking datafusion-expr-common v55.0.0 -> v55.0.0 (no change; assume patch)
     Checked [   0.278s] 223 checks: 223 pass, 31 skip
     Summary no semver update required
    Finished [  33.516s] datafusion-expr-common
    Building datafusion-functions-aggregate v55.0.0 (current)
       Built [  26.309s] (current)
     Parsing datafusion-functions-aggregate v55.0.0 (current)
      Parsed [   0.041s] (current)
    Building datafusion-functions-aggregate v55.0.0 (baseline)
       Built [  26.364s] (baseline)
     Parsing datafusion-functions-aggregate v55.0.0 (baseline)
      Parsed [   0.038s] (baseline)
    Checking datafusion-functions-aggregate v55.0.0 -> v55.0.0 (no change; assume patch)
     Checked [   0.236s] 223 checks: 223 pass, 31 skip
     Summary no semver update required
    Finished [  54.010s] datafusion-functions-aggregate
    Building datafusion-functions-aggregate-common v55.0.0 (current)
       Built [  17.297s] (current)
     Parsing datafusion-functions-aggregate-common v55.0.0 (current)
      Parsed [   0.018s] (current)
    Building datafusion-functions-aggregate-common v55.0.0 (baseline)
       Built [  17.229s] (baseline)
     Parsing datafusion-functions-aggregate-common v55.0.0 (baseline)
      Parsed [   0.016s] (baseline)
    Checking datafusion-functions-aggregate-common v55.0.0 -> v55.0.0 (no change; assume patch)
     Checked [   0.175s] 223 checks: 222 pass, 1 fail, 0 warn, 31 skip

--- failure function_requires_different_generic_type_params: function now requires a different number of generic type parameters ---

Description:
A function now requires a different number of generic type parameters than it used to. Uses of this function that supplied the previous number of generic types (e.g. via turbofish syntax) will be broken.
        ref: https://doc.rust-lang.org/reference/items/generics.html
       impl: https://github.com/obi1kenobi/cargo-semver-checks/tree/v0.50.0/src/lints/function_requires_different_generic_type_params.ron

Failed in:
  function accumulate_multiple (2 -> 3 generic types) in /home/runner/work/datafusion/datafusion/datafusion/functions-aggregate-common/src/aggregate/groups_accumulator/accumulate.rs:854
  function accumulate (2 -> 3 generic types) in /home/runner/work/datafusion/datafusion/datafusion/functions-aggregate-common/src/aggregate/groups_accumulator/accumulate.rs:739
  function accumulate_indices (1 -> 2 generic types) in /home/runner/work/datafusion/datafusion/datafusion/functions-aggregate-common/src/aggregate/groups_accumulator/accumulate.rs:910

     Summary semver requires new major version: 1 major and 0 minor checks failed
    Finished [  35.534s] datafusion-functions-aggregate-common
    Building datafusion-physical-expr v55.0.0 (current)
       Built [  24.342s] (current)
     Parsing datafusion-physical-expr v55.0.0 (current)
      Parsed [   0.042s] (current)
    Building datafusion-physical-expr v55.0.0 (baseline)
       Built [  24.684s] (baseline)
     Parsing datafusion-physical-expr v55.0.0 (baseline)
      Parsed [   0.043s] (baseline)
    Checking datafusion-physical-expr v55.0.0 -> v55.0.0 (no change; assume patch)
     Checked [   0.457s] 223 checks: 223 pass, 31 skip
     Summary no semver update required
    Finished [  50.558s] datafusion-physical-expr
    Building datafusion-physical-expr-common v55.0.0 (current)
       Built [  20.769s] (current)
     Parsing datafusion-physical-expr-common v55.0.0 (current)
      Parsed [   0.019s] (current)
    Building datafusion-physical-expr-common v55.0.0 (baseline)
       Built [  21.084s] (baseline)
     Parsing datafusion-physical-expr-common v55.0.0 (baseline)
      Parsed [   0.019s] (baseline)
    Checking datafusion-physical-expr-common v55.0.0 -> v55.0.0 (no change; assume patch)
     Checked [   0.297s] 223 checks: 223 pass, 31 skip
     Summary no semver update required
    Finished [  42.960s] datafusion-physical-expr-common
    Building datafusion-physical-plan v55.0.0 (current)
       Built [  32.499s] (current)
     Parsing datafusion-physical-plan v55.0.0 (current)
      Parsed [   0.162s] (current)
    Building datafusion-physical-plan v55.0.0 (baseline)
       Built [  32.852s] (baseline)
     Parsing datafusion-physical-plan v55.0.0 (baseline)
      Parsed [   0.132s] (baseline)
    Checking datafusion-physical-plan v55.0.0 -> v55.0.0 (no change; assume patch)
     Checked [   0.888s] 223 checks: 223 pass, 31 skip
     Summary no semver update required
    Finished [  67.765s] datafusion-physical-plan

adriangbot · 2026-09-03T17:14:19Z

🤖 Benchmark completed (GKE) | trigger

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

Comparing add-blocks-impl-from-scratch (d9d288b) to d2b626c (merge-base) diff

Run configuration

run benchmark tpch

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 add-blocks-impl-from-scratch
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query     ┃     HEAD ┃ add-blocks-impl-from-scratch ┃       Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 1  │ 38.17 ms │                     42.06 ms │ 1.10x slower │
│ QQuery 2  │ 18.54 ms │                     18.60 ms │    no change │
│ QQuery 3  │ 28.52 ms │                     28.51 ms │    no change │
│ QQuery 4  │ 17.57 ms │                     17.81 ms │    no change │
│ QQuery 5  │ 35.50 ms │                     35.91 ms │    no change │
│ QQuery 6  │ 16.16 ms │                     16.06 ms │    no change │
│ QQuery 7  │ 40.29 ms │                     40.78 ms │    no change │
│ QQuery 8  │ 40.26 ms │                     40.82 ms │    no change │
│ QQuery 9  │ 49.32 ms │                     48.68 ms │    no change │
│ QQuery 10 │ 41.83 ms │                     41.88 ms │    no change │
│ QQuery 11 │ 13.17 ms │                     13.60 ms │    no change │
│ QQuery 12 │ 24.00 ms │                     24.24 ms │    no change │
│ QQuery 13 │ 39.78 ms │                     42.18 ms │ 1.06x slower │
│ QQuery 14 │ 23.98 ms │                     24.13 ms │    no change │
│ QQuery 15 │ 30.07 ms │                     31.02 ms │    no change │
│ QQuery 16 │ 13.70 ms │                     14.36 ms │    no change │
│ QQuery 17 │ 70.76 ms │                     83.32 ms │ 1.18x slower │
│ QQuery 18 │ 59.22 ms │                     62.65 ms │ 1.06x slower │
│ QQuery 19 │ 32.22 ms │                     33.34 ms │    no change │
│ QQuery 20 │ 31.02 ms │                     34.67 ms │ 1.12x slower │
│ QQuery 21 │ 55.56 ms │                     55.34 ms │    no change │
│ QQuery 22 │ 13.88 ms │                     14.14 ms │    no change │
└───────────┴──────────┴──────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                           ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 733.51ms │
│ Total Time (add-blocks-impl-from-scratch)   │ 764.10ms │
│ Average Time (HEAD)                         │  33.34ms │
│ Average Time (add-blocks-impl-from-scratch) │  34.73ms │
│ Queries Faster                              │        0 │
│ Queries Slower                              │        5 │
│ Queries with No Change                      │       17 │
│ Queries with Failure                        │        0 │
└─────────────────────────────────────────────┴──────────┘

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

Comparing HEAD and add-blocks-impl-from-scratch
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃   add-blocks-impl-from-scratch ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 38.17 / 39.73 ±1.63 / 42.21 ms │ 42.06 / 43.52 ±1.69 / 45.86 ms │  1.10x slower │
│ QQuery 2  │ 18.54 / 19.04 ±0.68 / 20.38 ms │ 18.60 / 19.05 ±0.25 / 19.32 ms │     no change │
│ QQuery 3  │ 28.52 / 28.69 ±0.15 / 28.88 ms │ 28.51 / 28.75 ±0.19 / 28.99 ms │     no change │
│ QQuery 4  │ 17.57 / 17.70 ±0.10 / 17.88 ms │ 17.81 / 18.06 ±0.41 / 18.88 ms │     no change │
│ QQuery 5  │ 35.50 / 36.17 ±0.65 / 36.99 ms │ 35.91 / 36.56 ±0.92 / 38.39 ms │     no change │
│ QQuery 6  │ 16.16 / 17.19 ±0.92 / 18.33 ms │ 16.06 / 16.18 ±0.11 / 16.37 ms │ +1.06x faster │
│ QQuery 7  │ 40.29 / 42.11 ±1.57 / 44.75 ms │ 40.78 / 43.11 ±1.52 / 45.03 ms │     no change │
│ QQuery 8  │ 40.26 / 40.96 ±0.75 / 42.43 ms │ 40.82 / 41.62 ±0.99 / 43.50 ms │     no change │
│ QQuery 9  │ 49.32 / 49.99 ±0.66 / 51.16 ms │ 48.68 / 50.59 ±1.29 / 52.07 ms │     no change │
│ QQuery 10 │ 41.83 / 42.58 ±1.01 / 44.58 ms │ 41.88 / 42.29 ±0.44 / 43.10 ms │     no change │
│ QQuery 11 │ 13.17 / 13.40 ±0.17 / 13.66 ms │ 13.60 / 13.72 ±0.11 / 13.91 ms │     no change │
│ QQuery 12 │ 24.00 / 24.26 ±0.27 / 24.70 ms │ 24.24 / 24.46 ±0.18 / 24.73 ms │     no change │
│ QQuery 13 │ 39.78 / 40.39 ±0.57 / 41.30 ms │ 42.18 / 43.16 ±1.36 / 45.77 ms │  1.07x slower │
│ QQuery 14 │ 23.98 / 24.13 ±0.09 / 24.27 ms │ 24.13 / 24.37 ±0.16 / 24.63 ms │     no change │
│ QQuery 15 │ 30.07 / 30.26 ±0.13 / 30.40 ms │ 31.02 / 31.17 ±0.22 / 31.61 ms │     no change │
│ QQuery 16 │ 13.70 / 13.84 ±0.11 / 14.04 ms │ 14.36 / 14.48 ±0.07 / 14.57 ms │     no change │
│ QQuery 17 │ 70.76 / 71.35 ±0.38 / 71.74 ms │ 83.32 / 86.94 ±2.77 / 91.71 ms │  1.22x slower │
│ QQuery 18 │ 59.22 / 61.24 ±1.80 / 63.85 ms │ 62.65 / 63.48 ±0.71 / 64.73 ms │     no change │
│ QQuery 19 │ 32.22 / 32.42 ±0.16 / 32.69 ms │ 33.34 / 34.54 ±1.05 / 35.81 ms │  1.07x slower │
│ QQuery 20 │ 31.02 / 31.37 ±0.32 / 31.83 ms │ 34.67 / 35.42 ±0.41 / 35.83 ms │  1.13x slower │
│ QQuery 21 │ 55.56 / 56.30 ±0.56 / 57.19 ms │ 55.34 / 56.06 ±0.60 / 57.16 ms │     no change │
│ QQuery 22 │ 13.88 / 14.02 ±0.14 / 14.24 ms │ 14.14 / 14.27 ±0.11 / 14.46 ms │     no change │
└───────────┴────────────────────────────────┴────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                           ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 747.14ms │
│ Total Time (add-blocks-impl-from-scratch)   │ 781.81ms │
│ Average Time (HEAD)                         │  33.96ms │
│ Average Time (add-blocks-impl-from-scratch) │  35.54ms │
│ Queries Faster                              │        1 │
│ Queries Slower                              │        5 │
│ Queries with No Change                      │       16 │
│ Queries with Failure                        │        0 │
└─────────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric	Value
Wall time	5.0s
Peak memory	1.2 GiB
Avg memory	500.9 MiB
CPU user	20.9s
CPU sys	1.6s
Peak spill	0 B

tpch — branch

Metric	Value
Wall time	5.0s
Peak memory	1.3 GiB
Avg memory	521.6 MiB
CPU user	22.6s
CPU sys	1.6s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-09-03T17:15:52Z

🤖 Benchmark completed (GKE) | trigger

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

Comparing add-blocks-impl-from-scratch (d9d288b) to d2b626c (merge-base) diff

Run configuration

run benchmark tpcds

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 add-blocks-impl-from-scratch
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ add-blocks-impl-from-scratch ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │    5.68 ms │                      5.72 ms │     no change │
│ QQuery 2  │   82.06 ms │                     86.11 ms │     no change │
│ QQuery 3  │   29.05 ms │                     29.23 ms │     no change │
│ QQuery 4  │  500.23 ms │                    503.60 ms │     no change │
│ QQuery 5  │   52.13 ms │                     52.10 ms │     no change │
│ QQuery 6  │   36.81 ms │                     36.73 ms │     no change │
│ QQuery 7  │   75.37 ms │                     74.99 ms │     no change │
│ QQuery 8  │   37.10 ms │                     37.29 ms │     no change │
│ QQuery 9  │   52.67 ms │                     52.84 ms │     no change │
│ QQuery 10 │   63.36 ms │                     63.71 ms │     no change │
│ QQuery 11 │  314.73 ms │                    314.70 ms │     no change │
│ QQuery 12 │   28.92 ms │                     29.07 ms │     no change │
│ QQuery 13 │  118.60 ms │                    118.93 ms │     no change │
│ QQuery 14 │  418.11 ms │                    420.04 ms │     no change │
│ QQuery 15 │   58.61 ms │                     57.47 ms │     no change │
│ QQuery 16 │    6.80 ms │                      7.17 ms │  1.05x slower │
│ QQuery 17 │   79.99 ms │                     79.48 ms │     no change │
│ QQuery 18 │  105.39 ms │                    104.46 ms │     no change │
│ QQuery 19 │   41.34 ms │                     41.59 ms │     no change │
│ QQuery 20 │   35.61 ms │                     36.33 ms │     no change │
│ QQuery 21 │   17.67 ms │                     17.83 ms │     no change │
│ QQuery 22 │   63.78 ms │                     62.54 ms │     no change │
│ QQuery 23 │  311.25 ms │                    378.61 ms │  1.22x slower │
│ QQuery 24 │  198.61 ms │                    202.14 ms │     no change │
│ QQuery 25 │  110.35 ms │                    109.87 ms │     no change │
│ QQuery 26 │   49.66 ms │                     49.69 ms │     no change │
│ QQuery 27 │    6.47 ms │                      6.58 ms │     no change │
│ QQuery 28 │   60.97 ms │                     56.94 ms │ +1.07x faster │
│ QQuery 29 │   97.22 ms │                     98.12 ms │     no change │
│ QQuery 30 │   32.78 ms │                     33.21 ms │     no change │
│ QQuery 31 │  112.16 ms │                    112.73 ms │     no change │
│ QQuery 32 │   20.54 ms │                     21.32 ms │     no change │
│ QQuery 33 │   38.69 ms │                     38.65 ms │     no change │
│ QQuery 34 │   10.06 ms │                     10.03 ms │     no change │
│ QQuery 35 │   74.09 ms │                     74.58 ms │     no change │
│ QQuery 36 │    5.98 ms │                      5.89 ms │     no change │
│ QQuery 37 │    6.89 ms │                      6.90 ms │     no change │
│ QQuery 38 │   62.05 ms │                     66.02 ms │  1.06x slower │
│ QQuery 39 │   91.15 ms │                     93.55 ms │     no change │
│ QQuery 40 │   25.17 ms │                     24.21 ms │     no change │
│ QQuery 41 │   11.60 ms │                     11.20 ms │     no change │
│ QQuery 42 │   23.92 ms │                     23.85 ms │     no change │
│ QQuery 43 │    5.34 ms │                      5.50 ms │     no change │
│ QQuery 44 │    9.78 ms │                      9.65 ms │     no change │
│ QQuery 45 │   38.80 ms │                     39.60 ms │     no change │
│ QQuery 46 │   12.00 ms │                     12.08 ms │     no change │
│ QQuery 47 │  235.04 ms │                    244.73 ms │     no change │
│ QQuery 48 │   95.93 ms │                     95.65 ms │     no change │
│ QQuery 49 │   71.94 ms │                     72.04 ms │     no change │
│ QQuery 50 │   59.28 ms │                     59.34 ms │     no change │
│ QQuery 51 │   93.38 ms │                    100.58 ms │  1.08x slower │
│ QQuery 52 │   24.37 ms │                     24.46 ms │     no change │
│ QQuery 53 │   29.44 ms │                     29.91 ms │     no change │
│ QQuery 54 │   54.43 ms │                     55.27 ms │     no change │
│ QQuery 55 │   23.43 ms │                     23.38 ms │     no change │
│ QQuery 56 │   39.40 ms │                     39.34 ms │     no change │
│ QQuery 57 │  176.10 ms │                    183.49 ms │     no change │
│ QQuery 58 │  112.07 ms │                    113.60 ms │     no change │
│ QQuery 59 │  118.97 ms │                    128.40 ms │  1.08x slower │
│ QQuery 60 │   39.91 ms │                     39.51 ms │     no change │
│ QQuery 61 │   12.36 ms │                     12.24 ms │     no change │
│ QQuery 62 │   46.28 ms │                     47.84 ms │     no change │
│ QQuery 63 │   29.45 ms │                     29.40 ms │     no change │
│ QQuery 64 │  367.64 ms │                    369.44 ms │     no change │
│ QQuery 65 │  120.03 ms │                    129.04 ms │  1.08x slower │
│ QQuery 66 │   81.43 ms │                     83.22 ms │     no change │
│ QQuery 67 │  251.39 ms │                    256.47 ms │     no change │
│ QQuery 68 │   12.36 ms │                     11.97 ms │     no change │
│ QQuery 69 │   57.44 ms │                     57.16 ms │     no change │
│ QQuery 70 │  107.12 ms │                    107.87 ms │     no change │
│ QQuery 71 │   36.37 ms │                     35.51 ms │     no change │
│ QQuery 72 │ 1805.87 ms │                   1858.46 ms │     no change │
│ QQuery 73 │    9.81 ms │                     10.32 ms │  1.05x slower │
│ QQuery 74 │  171.81 ms │                    179.89 ms │     no change │
│ QQuery 75 │  146.19 ms │                    147.81 ms │     no change │
│ QQuery 76 │   35.41 ms │                     35.62 ms │     no change │
│ QQuery 77 │   61.69 ms │                     62.68 ms │     no change │
│ QQuery 78 │  218.68 ms │                    224.54 ms │     no change │
│ QQuery 79 │   66.60 ms │                     68.26 ms │     no change │
│ QQuery 80 │   98.38 ms │                     99.73 ms │     no change │
│ QQuery 81 │   25.93 ms │                     26.57 ms │     no change │
│ QQuery 82 │   16.49 ms │                     16.56 ms │     no change │
│ QQuery 83 │   34.43 ms │                     34.29 ms │     no change │
│ QQuery 84 │   29.88 ms │                     30.47 ms │     no change │
│ QQuery 85 │  103.83 ms │                    103.06 ms │     no change │
│ QQuery 86 │   25.52 ms │                     25.74 ms │     no change │
│ QQuery 87 │   61.95 ms │                     66.17 ms │  1.07x slower │
│ QQuery 88 │   64.36 ms │                     63.76 ms │     no change │
│ QQuery 89 │   35.52 ms │                     35.48 ms │     no change │
│ QQuery 90 │   17.39 ms │                     17.05 ms │     no change │
│ QQuery 91 │   45.50 ms │                     45.01 ms │     no change │
│ QQuery 92 │   30.01 ms │                     30.70 ms │     no change │
│ QQuery 93 │   49.52 ms │                     50.12 ms │     no change │
│ QQuery 94 │   37.62 ms │                     38.37 ms │     no change │
│ QQuery 95 │   80.52 ms │                     79.93 ms │     no change │
│ QQuery 96 │   24.24 ms │                     24.32 ms │     no change │
│ QQuery 97 │   52.82 ms │                     65.83 ms │  1.25x slower │
│ QQuery 98 │   42.94 ms │                     42.92 ms │     no change │
│ QQuery 99 │   69.96 ms │                     72.30 ms │     no change │
└───────────┴────────────┴──────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                           ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 9397.95ms │
│ Total Time (add-blocks-impl-from-scratch)   │ 9628.65ms │
│ Average Time (HEAD)                         │   94.93ms │
│ Average Time (add-blocks-impl-from-scratch) │   97.26ms │
│ Queries Faster                              │         1 │
│ Queries Slower                              │         9 │
│ Queries with No Change                      │        89 │
│ Queries with Failure                        │         0 │
└─────────────────────────────────────────────┴───────────┘

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

Comparing HEAD and add-blocks-impl-from-scratch
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃          add-blocks-impl-from-scratch ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.68 / 6.22 ±0.96 / 8.13 ms │           5.72 / 6.27 ±0.89 / 8.04 ms │     no change │
│ QQuery 2  │        82.06 / 82.38 ±0.26 / 82.75 ms │        86.11 / 86.36 ±0.23 / 86.78 ms │     no change │
│ QQuery 3  │        29.05 / 29.44 ±0.21 / 29.66 ms │        29.23 / 29.55 ±0.23 / 29.92 ms │     no change │
│ QQuery 4  │     500.23 / 503.72 ±5.52 / 514.70 ms │     503.60 / 508.72 ±4.09 / 514.96 ms │     no change │
│ QQuery 5  │        52.13 / 52.61 ±0.66 / 53.90 ms │        52.10 / 52.94 ±0.63 / 53.96 ms │     no change │
│ QQuery 6  │        36.81 / 37.26 ±0.35 / 37.67 ms │        36.73 / 36.91 ±0.17 / 37.20 ms │     no change │
│ QQuery 7  │        75.37 / 75.82 ±0.25 / 76.10 ms │        74.99 / 75.65 ±0.42 / 76.02 ms │     no change │
│ QQuery 8  │        37.10 / 38.89 ±2.15 / 43.04 ms │        37.29 / 38.13 ±1.04 / 40.16 ms │     no change │
│ QQuery 9  │        52.67 / 55.45 ±1.77 / 58.12 ms │        52.84 / 53.82 ±0.78 / 55.02 ms │     no change │
│ QQuery 10 │        63.36 / 63.80 ±0.26 / 64.05 ms │        63.71 / 63.95 ±0.18 / 64.25 ms │     no change │
│ QQuery 11 │     314.73 / 318.08 ±3.66 / 324.90 ms │     314.70 / 318.89 ±3.17 / 322.36 ms │     no change │
│ QQuery 12 │        28.92 / 29.17 ±0.21 / 29.43 ms │        29.07 / 29.56 ±0.39 / 30.12 ms │     no change │
│ QQuery 13 │     118.60 / 119.41 ±1.23 / 121.85 ms │     118.93 / 119.33 ±0.46 / 120.24 ms │     no change │
│ QQuery 14 │     418.11 / 419.39 ±0.93 / 421.01 ms │     420.04 / 425.29 ±6.24 / 437.29 ms │     no change │
│ QQuery 15 │        58.61 / 59.55 ±0.74 / 60.50 ms │        57.47 / 58.93 ±0.83 / 60.06 ms │     no change │
│ QQuery 16 │           6.80 / 7.01 ±0.20 / 7.37 ms │           7.17 / 7.32 ±0.13 / 7.48 ms │     no change │
│ QQuery 17 │        79.99 / 81.16 ±1.39 / 83.82 ms │        79.48 / 80.58 ±0.90 / 81.92 ms │     no change │
│ QQuery 18 │     105.39 / 107.26 ±1.95 / 110.83 ms │     104.46 / 105.89 ±0.92 / 107.16 ms │     no change │
│ QQuery 19 │        41.34 / 42.07 ±0.42 / 42.67 ms │        41.59 / 41.99 ±0.26 / 42.33 ms │     no change │
│ QQuery 20 │        35.61 / 35.88 ±0.24 / 36.19 ms │        36.33 / 37.31 ±1.18 / 38.95 ms │     no change │
│ QQuery 21 │        17.67 / 18.36 ±0.98 / 20.31 ms │        17.83 / 17.90 ±0.07 / 17.99 ms │     no change │
│ QQuery 22 │        63.78 / 66.98 ±3.13 / 71.22 ms │        62.54 / 63.61 ±0.65 / 64.33 ms │ +1.05x faster │
│ QQuery 23 │     311.25 / 315.81 ±3.26 / 320.22 ms │     378.61 / 383.71 ±4.29 / 391.22 ms │  1.22x slower │
│ QQuery 24 │     198.61 / 203.25 ±4.89 / 212.74 ms │     202.14 / 211.72 ±8.24 / 226.13 ms │     no change │
│ QQuery 25 │     110.35 / 113.59 ±3.50 / 119.65 ms │     109.87 / 111.40 ±1.76 / 114.77 ms │     no change │
│ QQuery 26 │        49.66 / 50.01 ±0.32 / 50.53 ms │        49.69 / 51.05 ±1.08 / 52.40 ms │     no change │
│ QQuery 27 │           6.47 / 6.68 ±0.22 / 7.07 ms │           6.58 / 6.86 ±0.18 / 7.13 ms │     no change │
│ QQuery 28 │        60.97 / 62.89 ±2.42 / 67.52 ms │        56.94 / 60.94 ±2.18 / 63.37 ms │     no change │
│ QQuery 29 │       97.22 / 98.60 ±1.56 / 101.52 ms │      98.12 / 101.12 ±2.89 / 106.47 ms │     no change │
│ QQuery 30 │        32.78 / 33.12 ±0.30 / 33.62 ms │        33.21 / 33.80 ±0.87 / 35.50 ms │     no change │
│ QQuery 31 │     112.16 / 114.78 ±2.64 / 118.73 ms │     112.73 / 113.76 ±1.32 / 116.32 ms │     no change │
│ QQuery 32 │        20.54 / 21.02 ±0.25 / 21.25 ms │        21.32 / 21.52 ±0.12 / 21.64 ms │     no change │
│ QQuery 33 │        38.69 / 38.83 ±0.11 / 39.00 ms │        38.65 / 38.95 ±0.36 / 39.52 ms │     no change │
│ QQuery 34 │        10.06 / 10.28 ±0.29 / 10.85 ms │        10.03 / 10.27 ±0.23 / 10.56 ms │     no change │
│ QQuery 35 │        74.09 / 75.70 ±1.47 / 78.28 ms │        74.58 / 75.62 ±0.83 / 76.71 ms │     no change │
│ QQuery 36 │           5.98 / 6.11 ±0.16 / 6.43 ms │           5.89 / 6.07 ±0.25 / 6.57 ms │     no change │
│ QQuery 37 │           6.89 / 7.02 ±0.08 / 7.14 ms │           6.90 / 6.99 ±0.14 / 7.26 ms │     no change │
│ QQuery 38 │        62.05 / 64.43 ±2.33 / 68.25 ms │        66.02 / 67.14 ±0.70 / 67.92 ms │     no change │
│ QQuery 39 │        91.15 / 91.87 ±0.79 / 93.30 ms │        93.55 / 95.22 ±1.13 / 96.77 ms │     no change │
│ QQuery 40 │        25.17 / 27.17 ±2.51 / 32.01 ms │        24.21 / 24.68 ±0.33 / 25.06 ms │ +1.10x faster │
│ QQuery 41 │        11.60 / 11.70 ±0.10 / 11.89 ms │        11.20 / 11.34 ±0.16 / 11.66 ms │     no change │
│ QQuery 42 │        23.92 / 24.91 ±0.97 / 26.72 ms │        23.85 / 24.25 ±0.24 / 24.53 ms │     no change │
│ QQuery 43 │           5.34 / 5.50 ±0.17 / 5.81 ms │           5.50 / 5.97 ±0.85 / 7.66 ms │  1.09x slower │
│ QQuery 44 │          9.78 / 9.95 ±0.15 / 10.17 ms │          9.65 / 9.91 ±0.21 / 10.14 ms │     no change │
│ QQuery 45 │        38.80 / 40.30 ±1.13 / 42.07 ms │        39.60 / 40.58 ±0.78 / 41.60 ms │     no change │
│ QQuery 46 │        12.00 / 12.30 ±0.21 / 12.57 ms │        12.08 / 13.60 ±2.10 / 17.67 ms │  1.11x slower │
│ QQuery 47 │     235.04 / 235.65 ±0.31 / 235.83 ms │     244.73 / 247.24 ±1.71 / 249.88 ms │     no change │
│ QQuery 48 │       95.93 / 99.85 ±3.85 / 106.68 ms │        95.65 / 95.99 ±0.22 / 96.31 ms │     no change │
│ QQuery 49 │        71.94 / 73.64 ±1.71 / 76.67 ms │        72.04 / 74.36 ±2.84 / 79.71 ms │     no change │
│ QQuery 50 │        59.28 / 59.57 ±0.15 / 59.73 ms │        59.34 / 60.71 ±1.28 / 63.12 ms │     no change │
│ QQuery 51 │        93.38 / 95.15 ±1.29 / 96.79 ms │     100.58 / 102.43 ±1.26 / 104.27 ms │  1.08x slower │
│ QQuery 52 │        24.37 / 24.60 ±0.14 / 24.77 ms │        24.46 / 25.69 ±2.17 / 30.02 ms │     no change │
│ QQuery 53 │        29.44 / 29.63 ±0.15 / 29.82 ms │        29.91 / 30.38 ±0.45 / 31.12 ms │     no change │
│ QQuery 54 │        54.43 / 54.93 ±0.40 / 55.54 ms │        55.27 / 55.57 ±0.25 / 56.01 ms │     no change │
│ QQuery 55 │        23.43 / 24.41 ±0.65 / 25.31 ms │        23.38 / 23.55 ±0.13 / 23.72 ms │     no change │
│ QQuery 56 │        39.40 / 40.40 ±1.17 / 42.68 ms │        39.34 / 39.91 ±0.57 / 40.99 ms │     no change │
│ QQuery 57 │     176.10 / 178.91 ±3.01 / 184.43 ms │     183.49 / 185.04 ±1.66 / 188.21 ms │     no change │
│ QQuery 58 │     112.07 / 113.22 ±0.70 / 114.26 ms │     113.60 / 115.08 ±1.12 / 116.58 ms │     no change │
│ QQuery 59 │     118.97 / 120.81 ±1.80 / 123.05 ms │     128.40 / 128.79 ±0.28 / 129.26 ms │  1.07x slower │
│ QQuery 60 │        39.91 / 40.33 ±0.28 / 40.72 ms │        39.51 / 40.05 ±0.40 / 40.51 ms │     no change │
│ QQuery 61 │        12.36 / 12.58 ±0.18 / 12.90 ms │        12.24 / 12.81 ±0.70 / 14.16 ms │     no change │
│ QQuery 62 │        46.28 / 46.68 ±0.34 / 47.07 ms │        47.84 / 48.07 ±0.23 / 48.46 ms │     no change │
│ QQuery 63 │        29.45 / 29.67 ±0.16 / 29.88 ms │        29.40 / 29.58 ±0.12 / 29.77 ms │     no change │
│ QQuery 64 │     367.64 / 373.17 ±6.31 / 385.05 ms │     369.44 / 374.12 ±4.55 / 381.62 ms │     no change │
│ QQuery 65 │     120.03 / 125.09 ±3.45 / 130.30 ms │     129.04 / 130.46 ±1.59 / 133.10 ms │     no change │
│ QQuery 66 │        81.43 / 81.80 ±0.30 / 82.30 ms │        83.22 / 84.35 ±1.21 / 86.54 ms │     no change │
│ QQuery 67 │     251.39 / 255.75 ±3.46 / 259.76 ms │     256.47 / 259.36 ±2.60 / 263.69 ms │     no change │
│ QQuery 68 │        12.36 / 12.53 ±0.13 / 12.70 ms │        11.97 / 12.34 ±0.25 / 12.63 ms │     no change │
│ QQuery 69 │        57.44 / 58.04 ±0.49 / 58.63 ms │        57.16 / 58.58 ±2.02 / 62.57 ms │     no change │
│ QQuery 70 │     107.12 / 107.72 ±0.52 / 108.64 ms │     107.87 / 108.89 ±1.20 / 111.25 ms │     no change │
│ QQuery 71 │        36.37 / 38.68 ±2.68 / 42.39 ms │        35.51 / 36.34 ±0.64 / 37.24 ms │ +1.06x faster │
│ QQuery 72 │ 1805.87 / 1872.71 ±39.32 / 1906.54 ms │ 1858.46 / 1913.25 ±28.09 / 1938.07 ms │     no change │
│ QQuery 73 │         9.81 / 10.02 ±0.15 / 10.27 ms │        10.32 / 10.70 ±0.29 / 11.11 ms │  1.07x slower │
│ QQuery 74 │     171.81 / 175.07 ±2.06 / 177.95 ms │     179.89 / 183.57 ±2.75 / 186.96 ms │     no change │
│ QQuery 75 │     146.19 / 149.03 ±2.17 / 152.33 ms │     147.81 / 149.41 ±1.33 / 151.38 ms │     no change │
│ QQuery 76 │        35.41 / 35.82 ±0.25 / 36.17 ms │        35.62 / 36.02 ±0.34 / 36.63 ms │     no change │
│ QQuery 77 │        61.69 / 62.22 ±0.29 / 62.55 ms │        62.68 / 62.99 ±0.39 / 63.65 ms │     no change │
│ QQuery 78 │     218.68 / 225.58 ±7.34 / 238.89 ms │     224.54 / 227.22 ±2.81 / 232.17 ms │     no change │
│ QQuery 79 │        66.60 / 68.79 ±2.28 / 72.86 ms │        68.26 / 71.24 ±3.38 / 77.74 ms │     no change │
│ QQuery 80 │      98.38 / 100.29 ±2.03 / 104.06 ms │      99.73 / 101.44 ±1.26 / 103.27 ms │     no change │
│ QQuery 81 │        25.93 / 26.14 ±0.15 / 26.38 ms │        26.57 / 26.89 ±0.27 / 27.30 ms │     no change │
│ QQuery 82 │        16.49 / 17.01 ±0.81 / 18.62 ms │        16.56 / 18.45 ±3.23 / 24.89 ms │  1.08x slower │
│ QQuery 83 │        34.43 / 35.73 ±2.44 / 40.62 ms │        34.29 / 34.95 ±0.37 / 35.41 ms │     no change │
│ QQuery 84 │        29.88 / 31.07 ±1.09 / 32.99 ms │        30.47 / 31.48 ±1.11 / 33.10 ms │     no change │
│ QQuery 85 │     103.83 / 105.01 ±0.87 / 106.15 ms │     103.06 / 104.39 ±0.95 / 105.99 ms │     no change │
│ QQuery 86 │        25.52 / 25.67 ±0.16 / 25.95 ms │        25.74 / 25.97 ±0.15 / 26.13 ms │     no change │
│ QQuery 87 │        61.95 / 64.44 ±1.62 / 66.77 ms │        66.17 / 68.74 ±2.82 / 73.41 ms │  1.07x slower │
│ QQuery 88 │        64.36 / 64.81 ±0.41 / 65.44 ms │        63.76 / 64.29 ±0.28 / 64.55 ms │     no change │
│ QQuery 89 │        35.52 / 35.70 ±0.22 / 36.11 ms │        35.48 / 36.00 ±0.28 / 36.23 ms │     no change │
│ QQuery 90 │        17.39 / 17.62 ±0.19 / 17.91 ms │        17.05 / 17.40 ±0.22 / 17.74 ms │     no change │
│ QQuery 91 │        45.50 / 45.66 ±0.11 / 45.86 ms │        45.01 / 46.93 ±2.72 / 52.28 ms │     no change │
│ QQuery 92 │        30.01 / 32.24 ±2.09 / 36.02 ms │        30.70 / 31.57 ±0.82 / 32.88 ms │     no change │
│ QQuery 93 │        49.52 / 51.23 ±1.20 / 53.13 ms │        50.12 / 51.18 ±1.14 / 53.29 ms │     no change │
│ QQuery 94 │        37.62 / 38.56 ±0.76 / 39.52 ms │        38.37 / 38.80 ±0.29 / 39.25 ms │     no change │
│ QQuery 95 │        80.52 / 81.25 ±0.47 / 81.78 ms │        79.93 / 81.09 ±0.83 / 82.46 ms │     no change │
│ QQuery 96 │        24.24 / 26.11 ±2.78 / 31.60 ms │        24.32 / 25.48 ±1.69 / 28.84 ms │     no change │
│ QQuery 97 │        52.82 / 54.22 ±1.49 / 57.04 ms │        65.83 / 67.41 ±1.58 / 69.97 ms │  1.24x slower │
│ QQuery 98 │        42.94 / 43.68 ±0.46 / 44.40 ms │        42.92 / 43.51 ±0.32 / 43.86 ms │     no change │
│ QQuery 99 │        69.96 / 70.46 ±0.31 / 70.88 ms │        72.30 / 73.85 ±2.27 / 78.34 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                           ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 9594.69ms │
│ Total Time (add-blocks-impl-from-scratch)   │ 9809.22ms │
│ Average Time (HEAD)                         │   96.92ms │
│ Average Time (add-blocks-impl-from-scratch) │   99.08ms │
│ Queries Faster                              │         3 │
│ Queries Slower                              │         9 │
│ Queries with No Change                      │        87 │
│ Queries with Failure                        │         0 │
└─────────────────────────────────────────────┴───────────┘

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	50.0s
Peak memory	2.3 GiB
Avg memory	1.5 GiB
CPU user	204.2s
CPU sys	5.6s
Peak spill	0 B

tpcds — branch

Metric	Value
Wall time	50.0s
Peak memory	2.5 GiB
Avg memory	1.8 GiB
CPU user	214.0s
CPU sys	5.7s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-09-03T17:25:16Z

🤖 Benchmark completed (GKE) | trigger

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

Comparing add-blocks-impl-from-scratch (d9d288b) to d2b626c (merge-base) diff

Run configuration

run benchmark clickbench_partitioned

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 add-blocks-impl-from-scratch
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ add-blocks-impl-from-scratch ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │    1.24 ms │                      5.79 ms │  4.67x slower │
│ QQuery 1  │   12.05 ms │                     12.25 ms │     no change │
│ QQuery 2  │   36.48 ms │                     36.55 ms │     no change │
│ QQuery 3  │   33.63 ms │                     31.24 ms │ +1.08x faster │
│ QQuery 4  │  223.05 ms │                    287.53 ms │  1.29x slower │
│ QQuery 5  │  273.95 ms │                         FAIL │  incomparable │
│ QQuery 6  │    1.26 ms │                      5.46 ms │  4.33x slower │
│ QQuery 7  │   13.88 ms │                     13.63 ms │     no change │
│ QQuery 8  │  332.69 ms │                   2694.51 ms │  8.10x slower │
│ QQuery 9  │  448.33 ms │                    551.77 ms │  1.23x slower │
│ QQuery 10 │   69.70 ms │                     83.01 ms │  1.19x slower │
│ QQuery 11 │   81.23 ms │                     93.25 ms │  1.15x slower │
│ QQuery 12 │  265.40 ms │                         FAIL │  incomparable │
│ QQuery 13 │  383.32 ms │                         FAIL │  incomparable │
│ QQuery 14 │  303.93 ms │                    651.20 ms │  2.14x slower │
│ QQuery 15 │  290.65 ms │                    358.99 ms │  1.24x slower │
│ QQuery 16 │  637.77 ms │                   6327.54 ms │  9.92x slower │
│ QQuery 17 │  631.85 ms │                   6429.62 ms │ 10.18x slower │
│ QQuery 18 │ 1298.52 ms │                  29180.92 ms │ 22.47x slower │
│ QQuery 19 │   27.33 ms │                     27.45 ms │     no change │
│ QQuery 20 │  515.27 ms │                    514.58 ms │     no change │
│ QQuery 21 │  514.51 ms │                    514.28 ms │     no change │
│ QQuery 22 │  981.61 ms │                    987.78 ms │     no change │
│ QQuery 23 │ 3077.79 ms │                   3033.16 ms │     no change │
│ QQuery 24 │   41.99 ms │                     41.20 ms │     no change │
│ QQuery 25 │  111.36 ms │                    110.08 ms │     no change │
│ QQuery 26 │   42.48 ms │                     41.51 ms │     no change │
│ QQuery 27 │  516.82 ms │                    529.51 ms │     no change │
│ QQuery 28 │ 2914.13 ms │                         FAIL │  incomparable │
│ QQuery 29 │   40.92 ms │                     40.44 ms │     no change │
│ QQuery 30 │  306.79 ms │                    594.92 ms │  1.94x slower │
│ QQuery 31 │  282.33 ms │                    815.33 ms │  2.89x slower │
│ QQuery 32 │  935.84 ms │                  31195.90 ms │ 33.33x slower │
│ QQuery 33 │ 1454.72 ms │                         FAIL │  incomparable │
│ QQuery 34 │ 1484.87 ms │                         FAIL │  incomparable │
│ QQuery 35 │  288.01 ms │                    372.13 ms │  1.29x slower │
│ QQuery 36 │   70.68 ms │                         FAIL │  incomparable │
│ QQuery 37 │   35.65 ms │                         FAIL │  incomparable │
│ QQuery 38 │   40.92 ms │                     39.94 ms │     no change │
│ QQuery 39 │  142.48 ms │                    179.64 ms │  1.26x slower │
│ QQuery 40 │   14.30 ms │                     15.04 ms │  1.05x slower │
│ QQuery 41 │   13.86 ms │                     14.31 ms │     no change │
│ QQuery 42 │   13.35 ms │                     14.70 ms │  1.10x slower │
└───────────┴────────────┴──────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                           ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 12324.23ms │
│ Total Time (add-blocks-impl-from-scratch)   │ 85845.16ms │
│ Average Time (HEAD)                         │   352.12ms │
│ Average Time (add-blocks-impl-from-scratch) │  2452.72ms │
│ Queries Faster                              │          1 │
│ Queries Slower                              │         19 │
│ Queries with No Change                      │         15 │
│ Queries with Failure                        │          8 │
└─────────────────────────────────────────────┴────────────┘

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

Comparing HEAD and add-blocks-impl-from-scratch
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃              add-blocks-impl-from-scratch ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.24 / 4.04 ±5.49 / 15.02 ms │              5.79 / 8.69 ±5.47 / 19.64 ms │  2.15x slower │
│ QQuery 1  │        12.05 / 12.32 ±0.25 / 12.73 ms │            12.25 / 12.33 ±0.08 / 12.48 ms │     no change │
│ QQuery 2  │        36.48 / 37.35 ±0.72 / 38.69 ms │            36.55 / 36.84 ±0.25 / 37.19 ms │     no change │
│ QQuery 3  │        33.63 / 34.46 ±0.58 / 35.23 ms │            31.24 / 31.61 ±0.38 / 32.31 ms │ +1.09x faster │
│ QQuery 4  │     223.05 / 225.69 ±2.19 / 229.62 ms │         287.53 / 292.64 ±2.97 / 296.57 ms │  1.30x slower │
│ QQuery 5  │     273.95 / 277.41 ±3.51 / 283.74 ms │                                      FAIL │  incomparable │
│ QQuery 6  │           1.26 / 1.40 ±0.22 / 1.85 ms │               5.46 / 6.89 ±0.88 / 7.74 ms │  4.91x slower │
│ QQuery 7  │        13.88 / 14.16 ±0.28 / 14.58 ms │            13.63 / 13.79 ±0.13 / 13.93 ms │     no change │
│ QQuery 8  │     332.69 / 341.44 ±8.41 / 355.80 ms │     2694.51 / 2796.59 ±59.02 / 2870.92 ms │  8.19x slower │
│ QQuery 9  │     448.33 / 460.56 ±8.17 / 469.19 ms │        551.77 / 567.63 ±11.77 / 588.02 ms │  1.23x slower │
│ QQuery 10 │        69.70 / 70.88 ±1.32 / 73.32 ms │            83.01 / 84.42 ±1.67 / 87.50 ms │  1.19x slower │
│ QQuery 11 │        81.23 / 82.42 ±0.92 / 83.98 ms │            93.25 / 95.20 ±1.22 / 97.09 ms │  1.16x slower │
│ QQuery 12 │     265.40 / 274.22 ±5.16 / 281.22 ms │                                      FAIL │  incomparable │
│ QQuery 13 │     383.32 / 396.73 ±7.68 / 406.49 ms │                                      FAIL │  incomparable │
│ QQuery 14 │     303.93 / 308.78 ±4.70 / 314.50 ms │        651.20 / 665.03 ±12.97 / 689.03 ms │  2.15x slower │
│ QQuery 15 │     290.65 / 299.12 ±7.05 / 308.01 ms │        358.99 / 369.91 ±10.57 / 389.20 ms │  1.24x slower │
│ QQuery 16 │    637.77 / 651.30 ±19.13 / 688.60 ms │     6327.54 / 6485.14 ±86.43 / 6565.47 ms │  9.96x slower │
│ QQuery 17 │    631.85 / 647.62 ±13.24 / 664.33 ms │    6429.62 / 6587.18 ±111.27 / 6764.60 ms │ 10.17x slower │
│ QQuery 18 │ 1298.52 / 1310.96 ±10.87 / 1326.29 ms │ 29180.92 / 30154.92 ±608.82 / 30767.57 ms │ 23.00x slower │
│ QQuery 19 │        27.33 / 31.22 ±7.44 / 46.09 ms │            27.45 / 27.90 ±0.26 / 28.18 ms │ +1.12x faster │
│ QQuery 20 │    515.27 / 529.75 ±12.63 / 545.20 ms │        514.58 / 530.63 ±12.14 / 552.08 ms │     no change │
│ QQuery 21 │     514.51 / 519.16 ±4.58 / 527.96 ms │         514.28 / 524.64 ±5.55 / 530.07 ms │     no change │
│ QQuery 22 │    981.61 / 994.37 ±8.63 / 1005.63 ms │         987.78 / 991.67 ±2.49 / 994.08 ms │     no change │
│ QQuery 23 │ 3077.79 / 3121.30 ±28.26 / 3155.60 ms │     3033.16 / 3059.48 ±24.15 / 3091.21 ms │     no change │
│ QQuery 24 │        41.99 / 45.90 ±6.66 / 59.17 ms │           41.20 / 53.83 ±11.46 / 70.62 ms │  1.17x slower │
│ QQuery 25 │     111.36 / 111.93 ±0.56 / 112.99 ms │         110.08 / 111.93 ±2.10 / 115.96 ms │     no change │
│ QQuery 26 │        42.48 / 47.96 ±8.33 / 64.36 ms │            41.51 / 42.39 ±0.99 / 44.28 ms │ +1.13x faster │
│ QQuery 27 │     516.82 / 523.80 ±6.00 / 532.69 ms │         529.51 / 539.78 ±8.11 / 552.44 ms │     no change │
│ QQuery 28 │ 2914.13 / 2940.29 ±18.32 / 2960.38 ms │                                      FAIL │  incomparable │
│ QQuery 29 │        40.92 / 42.97 ±3.37 / 49.67 ms │            40.44 / 43.56 ±3.32 / 48.90 ms │     no change │
│ QQuery 30 │    306.79 / 315.96 ±12.99 / 341.57 ms │         594.92 / 598.61 ±3.51 / 604.88 ms │  1.89x slower │
│ QQuery 31 │    282.33 / 301.20 ±20.09 / 337.96 ms │         815.33 / 823.84 ±8.78 / 837.96 ms │  2.74x slower │
│ QQuery 32 │    935.84 / 951.23 ±14.28 / 978.00 ms │  31195.90 / 31269.12 ±62.33 / 31363.96 ms │ 32.87x slower │
│ QQuery 33 │ 1454.72 / 1488.70 ±22.20 / 1514.77 ms │                                      FAIL │  incomparable │
│ QQuery 34 │ 1484.87 / 1525.97 ±34.51 / 1570.29 ms │                                      FAIL │  incomparable │
│ QQuery 35 │    288.01 / 321.38 ±39.31 / 386.79 ms │       372.13 / 467.14 ±176.64 / 820.13 ms │  1.45x slower │
│ QQuery 36 │      70.68 / 87.48 ±13.38 / 103.98 ms │                                      FAIL │  incomparable │
│ QQuery 37 │        35.65 / 37.07 ±1.40 / 39.61 ms │                                      FAIL │  incomparable │
│ QQuery 38 │        40.92 / 42.69 ±1.20 / 44.32 ms │            39.94 / 41.66 ±1.41 / 44.11 ms │     no change │
│ QQuery 39 │     142.48 / 153.33 ±5.60 / 157.62 ms │        179.64 / 196.93 ±21.01 / 236.81 ms │  1.28x slower │
│ QQuery 40 │        14.30 / 14.52 ±0.14 / 14.69 ms │            15.04 / 15.58 ±0.55 / 16.47 ms │  1.07x slower │
│ QQuery 41 │        13.86 / 14.00 ±0.09 / 14.12 ms │            14.31 / 14.44 ±0.11 / 14.62 ms │     no change │
│ QQuery 42 │        13.35 / 13.45 ±0.11 / 13.59 ms │            14.70 / 14.84 ±0.09 / 14.97 ms │  1.10x slower │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                           ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 12598.63ms │
│ Total Time (add-blocks-impl-from-scratch)   │ 87576.77ms │
│ Average Time (HEAD)                         │   359.96ms │
│ Average Time (add-blocks-impl-from-scratch) │  2502.19ms │
│ Queries Faster                              │          3 │
│ Queries Slower                              │         20 │
│ Queries with No Change                      │         12 │
│ Queries with Failure                        │          8 │
└─────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	100.0s
Peak memory	11.0 GiB
Avg memory	4.2 GiB
CPU user	1009.2s
CPU sys	69.1s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	440.1s
Peak memory	11.9 GiB
Avg memory	6.4 GiB
CPU user	4898.8s
CPU sys	64.6s
Peak spill	0 B

File an issue against this benchmark runner

rluvaton · 2026-09-03T22:27:58Z

run benchmarks

adriangbot · 2026-09-03T22:30:57Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5532954891-2127-66zt4 6.12.94+ #1 SMP Fri Jul 17 09:42:57 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 add-blocks-impl-from-scratch (5711b81) to d2b626c (merge-base) diff

Run configuration

run benchmark tpcds

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-09-03T22:31:02Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5532954891-2128-z4fh6 6.12.94+ #1 SMP Fri Jul 17 09:42:57 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 add-blocks-impl-from-scratch (5711b81) to d2b626c (merge-base) diff

Run configuration

run benchmark tpch

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-09-03T22:31:14Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5532954891-2126-w9jqh 6.12.94+ #1 SMP Fri Jul 17 09:42:57 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 add-blocks-impl-from-scratch (5711b81) to d2b626c (merge-base) diff

Run configuration

run benchmark clickbench_partitioned

Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-09-03T22:46:26Z

🤖 Benchmark completed (GKE) | trigger

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

Comparing add-blocks-impl-from-scratch (5711b81) to d2b626c (merge-base) diff

Run configuration

run benchmark tpch

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 add-blocks-impl-from-scratch
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃     HEAD ┃ add-blocks-impl-from-scratch ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 40.11 ms │                     43.76 ms │  1.09x slower │
│ QQuery 2  │ 19.63 ms │                     20.07 ms │     no change │
│ QQuery 3  │ 30.22 ms │                     30.42 ms │     no change │
│ QQuery 4  │ 18.38 ms │                     18.42 ms │     no change │
│ QQuery 5  │ 38.00 ms │                     35.35 ms │ +1.07x faster │
│ QQuery 6  │ 16.73 ms │                     15.85 ms │ +1.06x faster │
│ QQuery 7  │ 44.41 ms │                     41.13 ms │ +1.08x faster │
│ QQuery 8  │ 43.05 ms │                     40.34 ms │ +1.07x faster │
│ QQuery 9  │ 52.43 ms │                     48.54 ms │ +1.08x faster │
│ QQuery 10 │ 45.25 ms │                     42.01 ms │ +1.08x faster │
│ QQuery 11 │ 14.47 ms │                     13.98 ms │     no change │
│ QQuery 12 │ 25.21 ms │                     24.46 ms │     no change │
│ QQuery 13 │ 44.95 ms │                     42.41 ms │ +1.06x faster │
│ QQuery 14 │ 25.69 ms │                     24.21 ms │ +1.06x faster │
│ QQuery 15 │ 32.75 ms │                     31.11 ms │ +1.05x faster │
│ QQuery 16 │ 14.80 ms │                     14.12 ms │     no change │
│ QQuery 17 │ 79.88 ms │                     81.56 ms │     no change │
│ QQuery 18 │ 65.57 ms │                     60.42 ms │ +1.09x faster │
│ QQuery 19 │ 34.69 ms │                     32.37 ms │ +1.07x faster │
│ QQuery 20 │ 33.20 ms │                     31.96 ms │     no change │
│ QQuery 21 │ 61.42 ms │                     54.31 ms │ +1.13x faster │
│ QQuery 22 │ 15.54 ms │                     13.72 ms │ +1.13x faster │
└───────────┴──────────┴──────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                           ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 796.36ms │
│ Total Time (add-blocks-impl-from-scratch)   │ 760.52ms │
│ Average Time (HEAD)                         │  36.20ms │
│ Average Time (add-blocks-impl-from-scratch) │  34.57ms │
│ Queries Faster                              │       13 │
│ Queries Slower                              │        1 │
│ Queries with No Change                      │        8 │
│ Queries with Failure                        │        0 │
└─────────────────────────────────────────────┴──────────┘

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

Comparing HEAD and add-blocks-impl-from-scratch
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃   add-blocks-impl-from-scratch ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 40.11 / 41.91 ±1.49 / 43.52 ms │ 43.76 / 44.84 ±1.20 / 46.85 ms │  1.07x slower │
│ QQuery 2  │ 19.63 / 19.86 ±0.16 / 20.10 ms │ 20.07 / 20.30 ±0.15 / 20.49 ms │     no change │
│ QQuery 3  │ 30.22 / 30.62 ±0.23 / 30.91 ms │ 30.42 / 30.61 ±0.16 / 30.90 ms │     no change │
│ QQuery 4  │ 18.38 / 18.76 ±0.24 / 19.09 ms │ 18.42 / 18.96 ±0.77 / 20.49 ms │     no change │
│ QQuery 5  │ 38.00 / 38.45 ±0.30 / 38.84 ms │ 35.35 / 36.37 ±0.85 / 37.88 ms │ +1.06x faster │
│ QQuery 6  │ 16.73 / 17.11 ±0.39 / 17.84 ms │ 15.85 / 16.23 ±0.39 / 16.94 ms │ +1.05x faster │
│ QQuery 7  │ 44.41 / 45.96 ±0.94 / 47.15 ms │ 41.13 / 42.67 ±1.55 / 45.31 ms │ +1.08x faster │
│ QQuery 8  │ 43.05 / 43.22 ±0.16 / 43.51 ms │ 40.34 / 40.71 ±0.30 / 41.26 ms │ +1.06x faster │
│ QQuery 9  │ 52.43 / 53.78 ±1.45 / 55.81 ms │ 48.54 / 51.24 ±2.25 / 54.31 ms │     no change │
│ QQuery 10 │ 45.25 / 45.36 ±0.10 / 45.51 ms │ 42.01 / 43.61 ±1.59 / 46.41 ms │     no change │
│ QQuery 11 │ 14.47 / 14.93 ±0.43 / 15.71 ms │ 13.98 / 14.27 ±0.17 / 14.46 ms │     no change │
│ QQuery 12 │ 25.21 / 25.43 ±0.33 / 26.08 ms │ 24.46 / 25.40 ±1.34 / 28.05 ms │     no change │
│ QQuery 13 │ 44.95 / 45.26 ±0.23 / 45.66 ms │ 42.41 / 45.56 ±3.13 / 51.40 ms │     no change │
│ QQuery 14 │ 25.69 / 26.00 ±0.16 / 26.12 ms │ 24.21 / 24.25 ±0.04 / 24.31 ms │ +1.07x faster │
│ QQuery 15 │ 32.75 / 33.82 ±0.82 / 34.86 ms │ 31.11 / 31.98 ±0.59 / 32.67 ms │ +1.06x faster │
│ QQuery 16 │ 14.80 / 15.47 ±0.95 / 17.36 ms │ 14.12 / 14.33 ±0.15 / 14.52 ms │ +1.08x faster │
│ QQuery 17 │ 79.88 / 82.28 ±1.66 / 85.04 ms │ 81.56 / 82.19 ±0.40 / 82.83 ms │     no change │
│ QQuery 18 │ 65.57 / 66.88 ±1.86 / 70.56 ms │ 60.42 / 62.37 ±1.33 / 64.33 ms │ +1.07x faster │
│ QQuery 19 │ 34.69 / 35.74 ±1.15 / 37.85 ms │ 32.37 / 32.71 ±0.40 / 33.41 ms │ +1.09x faster │
│ QQuery 20 │ 33.20 / 33.97 ±1.10 / 36.16 ms │ 31.96 / 32.48 ±0.65 / 33.68 ms │     no change │
│ QQuery 21 │ 61.42 / 62.62 ±1.32 / 64.72 ms │ 54.31 / 55.09 ±0.82 / 56.41 ms │ +1.14x faster │
│ QQuery 22 │ 15.54 / 16.87 ±2.23 / 21.30 ms │ 13.72 / 14.06 ±0.20 / 14.32 ms │ +1.20x faster │
└───────────┴────────────────────────────────┴────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                           ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 814.29ms │
│ Total Time (add-blocks-impl-from-scratch)   │ 780.25ms │
│ Average Time (HEAD)                         │  37.01ms │
│ Average Time (add-blocks-impl-from-scratch) │  35.47ms │
│ Queries Faster                              │       11 │
│ Queries Slower                              │        1 │
│ Queries with No Change                      │       10 │
│ Queries with Failure                        │        0 │
└─────────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric	Value
Wall time	5.0s
Peak memory	1.1 GiB
Avg memory	634.3 MiB
CPU user	22.9s
CPU sys	2.0s
Peak spill	0 B

tpch — branch

Metric	Value
Wall time	5.0s
Peak memory	1.2 GiB
Avg memory	496.4 MiB
CPU user	22.3s
CPU sys	1.7s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-09-03T22:47:50Z

🤖 Benchmark completed (GKE) | trigger

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

Comparing add-blocks-impl-from-scratch (5711b81) to d2b626c (merge-base) diff

Run configuration

run benchmark tpcds

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 add-blocks-impl-from-scratch
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ add-blocks-impl-from-scratch ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │    5.62 ms │                      5.68 ms │     no change │
│ QQuery 2  │   81.21 ms │                     85.24 ms │     no change │
│ QQuery 3  │   28.91 ms │                     29.00 ms │     no change │
│ QQuery 4  │  483.63 ms │                    504.07 ms │     no change │
│ QQuery 5  │   51.70 ms │                     52.16 ms │     no change │
│ QQuery 6  │   35.92 ms │                     36.93 ms │     no change │
│ QQuery 7  │   74.77 ms │                     75.77 ms │     no change │
│ QQuery 8  │   36.28 ms │                     36.52 ms │     no change │
│ QQuery 9  │   52.27 ms │                     52.51 ms │     no change │
│ QQuery 10 │   62.25 ms │                     62.54 ms │     no change │
│ QQuery 11 │  302.24 ms │                    308.26 ms │     no change │
│ QQuery 12 │   28.82 ms │                     28.55 ms │     no change │
│ QQuery 13 │  118.38 ms │                    117.10 ms │     no change │
│ QQuery 14 │  414.69 ms │                    420.74 ms │     no change │
│ QQuery 15 │   60.09 ms │                     57.37 ms │     no change │
│ QQuery 16 │    7.01 ms │                      6.70 ms │     no change │
│ QQuery 17 │   80.68 ms │                     79.26 ms │     no change │
│ QQuery 18 │  106.79 ms │                    104.29 ms │     no change │
│ QQuery 19 │   41.99 ms │                     41.05 ms │     no change │
│ QQuery 20 │   36.02 ms │                     35.94 ms │     no change │
│ QQuery 21 │   17.44 ms │                     17.62 ms │     no change │
│ QQuery 22 │   63.49 ms │                     63.14 ms │     no change │
│ QQuery 23 │  309.12 ms │                    335.56 ms │  1.09x slower │
│ QQuery 24 │  195.14 ms │                    197.52 ms │     no change │
│ QQuery 25 │  110.52 ms │                    110.46 ms │     no change │
│ QQuery 26 │   48.49 ms │                     49.19 ms │     no change │
│ QQuery 27 │    6.07 ms │                      6.46 ms │  1.06x slower │
│ QQuery 28 │   58.34 ms │                     56.80 ms │     no change │
│ QQuery 29 │   97.58 ms │                     98.12 ms │     no change │
│ QQuery 30 │   32.51 ms │                     32.63 ms │     no change │
│ QQuery 31 │  111.12 ms │                    114.94 ms │     no change │
│ QQuery 32 │   20.23 ms │                     20.93 ms │     no change │
│ QQuery 33 │   38.11 ms │                     38.62 ms │     no change │
│ QQuery 34 │    9.99 ms │                      9.95 ms │     no change │
│ QQuery 35 │   72.89 ms │                     74.82 ms │     no change │
│ QQuery 36 │    5.82 ms │                      5.93 ms │     no change │
│ QQuery 37 │    6.96 ms │                      6.91 ms │     no change │
│ QQuery 38 │   62.48 ms │                     66.07 ms │  1.06x slower │
│ QQuery 39 │   88.97 ms │                    101.58 ms │  1.14x slower │
│ QQuery 40 │   23.99 ms │                     24.41 ms │     no change │
│ QQuery 41 │   10.99 ms │                     11.37 ms │     no change │
│ QQuery 42 │   23.83 ms │                     24.59 ms │     no change │
│ QQuery 43 │    5.19 ms │                      5.31 ms │     no change │
│ QQuery 44 │    9.18 ms │                      9.38 ms │     no change │
│ QQuery 45 │   39.71 ms │                     39.61 ms │     no change │
│ QQuery 46 │   11.70 ms │                     11.69 ms │     no change │
│ QQuery 47 │  226.81 ms │                    259.71 ms │  1.15x slower │
│ QQuery 48 │   95.58 ms │                     97.34 ms │     no change │
│ QQuery 49 │   71.35 ms │                     72.13 ms │     no change │
│ QQuery 50 │   59.00 ms │                     60.33 ms │     no change │
│ QQuery 51 │   92.12 ms │                     96.23 ms │     no change │
│ QQuery 52 │   25.13 ms │                     24.64 ms │     no change │
│ QQuery 53 │   29.97 ms │                     29.45 ms │     no change │
│ QQuery 54 │   56.13 ms │                     54.23 ms │     no change │
│ QQuery 55 │   23.76 ms │                     23.31 ms │     no change │
│ QQuery 56 │   39.93 ms │                     39.14 ms │     no change │
│ QQuery 57 │  177.15 ms │                    184.07 ms │     no change │
│ QQuery 58 │  113.56 ms │                    111.95 ms │     no change │
│ QQuery 59 │  119.42 ms │                    125.67 ms │  1.05x slower │
│ QQuery 60 │   39.70 ms │                     39.47 ms │     no change │
│ QQuery 61 │   12.25 ms │                     11.90 ms │     no change │
│ QQuery 62 │   45.87 ms │                     47.90 ms │     no change │
│ QQuery 63 │   28.92 ms │                     29.19 ms │     no change │
│ QQuery 64 │  363.52 ms │                    367.65 ms │     no change │
│ QQuery 65 │  121.32 ms │                    122.87 ms │     no change │
│ QQuery 66 │   82.18 ms │                     82.27 ms │     no change │
│ QQuery 67 │  259.37 ms │                    246.76 ms │     no change │
│ QQuery 68 │   12.37 ms │                     12.07 ms │     no change │
│ QQuery 69 │   57.39 ms │                     56.91 ms │     no change │
│ QQuery 70 │  107.92 ms │                    107.59 ms │     no change │
│ QQuery 71 │   34.82 ms │                     35.33 ms │     no change │
│ QQuery 72 │ 1834.66 ms │                   1776.08 ms │     no change │
│ QQuery 73 │   10.21 ms │                     10.20 ms │     no change │
│ QQuery 74 │  180.76 ms │                    176.94 ms │     no change │
│ QQuery 75 │  146.50 ms │                    149.10 ms │     no change │
│ QQuery 76 │   35.39 ms │                     35.69 ms │     no change │
│ QQuery 77 │   60.94 ms │                     61.32 ms │     no change │
│ QQuery 78 │  218.70 ms │                    206.06 ms │ +1.06x faster │
│ QQuery 79 │   66.73 ms │                     66.72 ms │     no change │
│ QQuery 80 │   98.99 ms │                     97.80 ms │     no change │
│ QQuery 81 │   26.07 ms │                     26.40 ms │     no change │
│ QQuery 82 │   16.05 ms │                     16.75 ms │     no change │
│ QQuery 83 │   34.15 ms │                     34.23 ms │     no change │
│ QQuery 84 │   29.32 ms │                     29.37 ms │     no change │
│ QQuery 85 │  102.39 ms │                    102.19 ms │     no change │
│ QQuery 86 │   25.11 ms │                     25.62 ms │     no change │
│ QQuery 87 │   61.75 ms │                     64.96 ms │  1.05x slower │
│ QQuery 88 │   63.26 ms │                     62.84 ms │     no change │
│ QQuery 89 │   35.39 ms │                     36.12 ms │     no change │
│ QQuery 90 │   16.88 ms │                     17.07 ms │     no change │
│ QQuery 91 │   44.55 ms │                     45.03 ms │     no change │
│ QQuery 92 │   29.63 ms │                     30.57 ms │     no change │
│ QQuery 93 │   49.59 ms │                     50.21 ms │     no change │
│ QQuery 94 │   38.41 ms │                     38.64 ms │     no change │
│ QQuery 95 │   80.01 ms │                     80.78 ms │     no change │
│ QQuery 96 │   23.88 ms │                     23.79 ms │     no change │
│ QQuery 97 │   52.09 ms │                     55.74 ms │  1.07x slower │
│ QQuery 98 │   42.65 ms │                     43.58 ms │     no change │
│ QQuery 99 │   69.56 ms │                     72.95 ms │     no change │
└───────────┴────────────┴──────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                           ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 9378.35ms │
│ Total Time (add-blocks-impl-from-scratch)   │ 9448.17ms │
│ Average Time (HEAD)                         │   94.73ms │
│ Average Time (add-blocks-impl-from-scratch) │   95.44ms │
│ Queries Faster                              │         1 │
│ Queries Slower                              │         8 │
│ Queries with No Change                      │        90 │
│ Queries with Failure                        │         0 │
└─────────────────────────────────────────────┴───────────┘

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

Comparing HEAD and add-blocks-impl-from-scratch
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃          add-blocks-impl-from-scratch ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.62 / 6.11 ±0.92 / 7.94 ms │           5.68 / 6.19 ±0.91 / 7.99 ms │     no change │
│ QQuery 2  │        81.21 / 81.52 ±0.19 / 81.73 ms │        85.24 / 85.34 ±0.09 / 85.48 ms │     no change │
│ QQuery 3  │        28.91 / 29.15 ±0.16 / 29.29 ms │        29.00 / 29.15 ±0.11 / 29.33 ms │     no change │
│ QQuery 4  │     483.63 / 487.27 ±3.90 / 494.31 ms │    504.07 / 530.18 ±13.59 / 541.06 ms │  1.09x slower │
│ QQuery 5  │        51.70 / 52.75 ±1.10 / 54.78 ms │        52.16 / 54.14 ±1.33 / 56.28 ms │     no change │
│ QQuery 6  │        35.92 / 36.41 ±0.31 / 36.76 ms │        36.93 / 37.39 ±0.35 / 37.89 ms │     no change │
│ QQuery 7  │        74.77 / 75.30 ±0.34 / 75.76 ms │        75.77 / 77.29 ±1.82 / 80.73 ms │     no change │
│ QQuery 8  │        36.28 / 37.72 ±2.52 / 42.76 ms │        36.52 / 36.95 ±0.34 / 37.32 ms │     no change │
│ QQuery 9  │        52.27 / 54.03 ±2.49 / 58.95 ms │        52.51 / 54.20 ±2.26 / 58.62 ms │     no change │
│ QQuery 10 │        62.25 / 63.04 ±0.48 / 63.76 ms │        62.54 / 62.95 ±0.24 / 63.22 ms │     no change │
│ QQuery 11 │     302.24 / 305.02 ±2.08 / 307.39 ms │     308.26 / 312.50 ±3.84 / 317.67 ms │     no change │
│ QQuery 12 │        28.82 / 29.34 ±0.32 / 29.84 ms │        28.55 / 28.76 ±0.20 / 29.01 ms │     no change │
│ QQuery 13 │     118.38 / 119.39 ±0.96 / 120.75 ms │     117.10 / 118.26 ±1.16 / 120.41 ms │     no change │
│ QQuery 14 │     414.69 / 420.53 ±4.71 / 426.69 ms │     420.74 / 425.32 ±3.74 / 431.39 ms │     no change │
│ QQuery 15 │        60.09 / 60.70 ±0.69 / 61.95 ms │        57.37 / 58.04 ±0.63 / 59.22 ms │     no change │
│ QQuery 16 │           7.01 / 7.15 ±0.19 / 7.52 ms │           6.70 / 6.89 ±0.24 / 7.36 ms │     no change │
│ QQuery 17 │        80.68 / 81.87 ±1.28 / 84.25 ms │        79.26 / 80.80 ±1.49 / 83.44 ms │     no change │
│ QQuery 18 │     106.79 / 107.56 ±0.64 / 108.61 ms │     104.29 / 104.94 ±0.54 / 105.92 ms │     no change │
│ QQuery 19 │        41.99 / 42.19 ±0.18 / 42.46 ms │        41.05 / 41.83 ±0.47 / 42.50 ms │     no change │
│ QQuery 20 │        36.02 / 36.34 ±0.20 / 36.56 ms │        35.94 / 36.63 ±0.98 / 38.56 ms │     no change │
│ QQuery 21 │        17.44 / 18.17 ±1.12 / 20.40 ms │        17.62 / 17.85 ±0.24 / 18.18 ms │     no change │
│ QQuery 22 │        63.49 / 63.99 ±0.37 / 64.51 ms │        63.14 / 64.21 ±0.68 / 64.97 ms │     no change │
│ QQuery 23 │     309.12 / 315.15 ±6.01 / 326.65 ms │     335.56 / 341.11 ±4.96 / 347.22 ms │  1.08x slower │
│ QQuery 24 │     195.14 / 200.63 ±3.91 / 206.88 ms │     197.52 / 201.05 ±4.99 / 210.88 ms │     no change │
│ QQuery 25 │     110.52 / 112.35 ±1.45 / 114.40 ms │     110.46 / 112.20 ±2.41 / 116.95 ms │     no change │
│ QQuery 26 │        48.49 / 48.87 ±0.23 / 49.16 ms │        49.19 / 49.39 ±0.22 / 49.75 ms │     no change │
│ QQuery 27 │           6.07 / 6.28 ±0.15 / 6.54 ms │           6.46 / 6.61 ±0.18 / 6.94 ms │  1.05x slower │
│ QQuery 28 │        58.34 / 60.20 ±0.95 / 60.85 ms │        56.80 / 63.21 ±3.84 / 68.12 ms │     no change │
│ QQuery 29 │       97.58 / 98.76 ±0.98 / 100.10 ms │       98.12 / 99.07 ±1.61 / 102.29 ms │     no change │
│ QQuery 30 │        32.51 / 32.76 ±0.36 / 33.47 ms │        32.63 / 33.21 ±0.33 / 33.53 ms │     no change │
│ QQuery 31 │     111.12 / 113.41 ±3.66 / 120.67 ms │     114.94 / 116.08 ±0.65 / 116.76 ms │     no change │
│ QQuery 32 │        20.23 / 20.76 ±0.30 / 21.14 ms │        20.93 / 21.14 ±0.21 / 21.52 ms │     no change │
│ QQuery 33 │        38.11 / 38.30 ±0.12 / 38.47 ms │        38.62 / 39.06 ±0.38 / 39.51 ms │     no change │
│ QQuery 34 │         9.99 / 10.26 ±0.19 / 10.55 ms │         9.95 / 10.27 ±0.29 / 10.81 ms │     no change │
│ QQuery 35 │        72.89 / 73.50 ±0.51 / 74.28 ms │        74.82 / 77.21 ±2.02 / 79.94 ms │  1.05x slower │
│ QQuery 36 │          5.82 / 7.15 ±2.33 / 11.78 ms │           5.93 / 6.10 ±0.20 / 6.49 ms │ +1.17x faster │
│ QQuery 37 │           6.96 / 7.12 ±0.20 / 7.42 ms │           6.91 / 7.05 ±0.11 / 7.24 ms │     no change │
│ QQuery 38 │        62.48 / 63.72 ±0.85 / 64.51 ms │        66.07 / 66.85 ±0.87 / 68.45 ms │     no change │
│ QQuery 39 │        88.97 / 89.81 ±1.14 / 92.08 ms │     101.58 / 103.74 ±2.99 / 109.58 ms │  1.16x slower │
│ QQuery 40 │        23.99 / 24.25 ±0.22 / 24.51 ms │        24.41 / 25.59 ±0.87 / 26.55 ms │  1.06x slower │
│ QQuery 41 │        10.99 / 11.27 ±0.29 / 11.81 ms │        11.37 / 11.48 ±0.15 / 11.76 ms │     no change │
│ QQuery 42 │        23.83 / 25.95 ±3.41 / 32.69 ms │        24.59 / 24.91 ±0.36 / 25.39 ms │     no change │
│ QQuery 43 │           5.19 / 5.32 ±0.16 / 5.63 ms │           5.31 / 5.43 ±0.20 / 5.82 ms │     no change │
│ QQuery 44 │           9.18 / 9.39 ±0.17 / 9.68 ms │           9.38 / 9.53 ±0.11 / 9.71 ms │     no change │
│ QQuery 45 │        39.71 / 40.09 ±0.33 / 40.58 ms │        39.61 / 39.93 ±0.24 / 40.25 ms │     no change │
│ QQuery 46 │        11.70 / 12.54 ±0.80 / 13.98 ms │        11.69 / 12.05 ±0.26 / 12.41 ms │     no change │
│ QQuery 47 │     226.81 / 229.63 ±2.31 / 232.33 ms │     259.71 / 262.81 ±2.82 / 267.77 ms │  1.14x slower │
│ QQuery 48 │       95.58 / 98.38 ±4.04 / 106.41 ms │        97.34 / 98.27 ±0.54 / 99.00 ms │     no change │
│ QQuery 49 │        71.35 / 72.34 ±0.87 / 73.64 ms │        72.13 / 72.97 ±0.59 / 73.66 ms │     no change │
│ QQuery 50 │        59.00 / 59.56 ±0.75 / 61.04 ms │        60.33 / 63.10 ±2.98 / 68.30 ms │  1.06x slower │
│ QQuery 51 │       92.12 / 96.06 ±2.98 / 100.28 ms │        96.23 / 97.21 ±0.91 / 98.74 ms │     no change │
│ QQuery 52 │        25.13 / 26.70 ±2.06 / 30.73 ms │        24.64 / 25.01 ±0.28 / 25.40 ms │ +1.07x faster │
│ QQuery 53 │        29.97 / 30.24 ±0.20 / 30.52 ms │        29.45 / 31.52 ±3.45 / 38.40 ms │     no change │
│ QQuery 54 │        56.13 / 56.31 ±0.20 / 56.65 ms │        54.23 / 54.79 ±0.54 / 55.52 ms │     no change │
│ QQuery 55 │        23.76 / 24.18 ±0.28 / 24.49 ms │        23.31 / 23.73 ±0.28 / 24.15 ms │     no change │
│ QQuery 56 │        39.93 / 40.89 ±0.90 / 42.54 ms │        39.14 / 39.59 ±0.46 / 40.21 ms │     no change │
│ QQuery 57 │     177.15 / 181.33 ±4.42 / 189.56 ms │     184.07 / 185.24 ±1.18 / 186.91 ms │     no change │
│ QQuery 58 │     113.56 / 116.10 ±3.85 / 123.75 ms │     111.95 / 114.36 ±2.18 / 118.38 ms │     no change │
│ QQuery 59 │     119.42 / 119.65 ±0.22 / 119.97 ms │     125.67 / 126.24 ±0.38 / 126.76 ms │  1.06x slower │
│ QQuery 60 │        39.70 / 40.30 ±0.54 / 41.08 ms │        39.47 / 40.03 ±0.31 / 40.34 ms │     no change │
│ QQuery 61 │        12.25 / 12.47 ±0.16 / 12.72 ms │        11.90 / 12.06 ±0.18 / 12.39 ms │     no change │
│ QQuery 62 │        45.87 / 46.22 ±0.22 / 46.50 ms │        47.90 / 49.64 ±2.51 / 54.49 ms │  1.07x slower │
│ QQuery 63 │        28.92 / 29.29 ±0.27 / 29.67 ms │        29.19 / 29.45 ±0.33 / 30.08 ms │     no change │
│ QQuery 64 │     363.52 / 369.87 ±6.08 / 380.99 ms │     367.65 / 371.84 ±3.90 / 378.63 ms │     no change │
│ QQuery 65 │     121.32 / 123.76 ±1.29 / 124.86 ms │     122.87 / 126.67 ±3.23 / 132.54 ms │     no change │
│ QQuery 66 │        82.18 / 83.44 ±0.99 / 85.21 ms │        82.27 / 86.48 ±6.65 / 99.72 ms │     no change │
│ QQuery 67 │     259.37 / 263.54 ±5.05 / 273.34 ms │     246.76 / 252.97 ±5.40 / 260.69 ms │     no change │
│ QQuery 68 │        12.37 / 12.90 ±0.40 / 13.33 ms │        12.07 / 12.20 ±0.13 / 12.44 ms │ +1.06x faster │
│ QQuery 69 │        57.39 / 58.10 ±0.66 / 59.14 ms │        56.91 / 57.30 ±0.22 / 57.54 ms │     no change │
│ QQuery 70 │     107.92 / 109.61 ±1.56 / 112.09 ms │     107.59 / 110.58 ±4.82 / 120.20 ms │     no change │
│ QQuery 71 │        34.82 / 36.80 ±2.15 / 40.81 ms │        35.33 / 36.08 ±1.09 / 38.18 ms │     no change │
│ QQuery 72 │ 1834.66 / 1888.14 ±50.48 / 1981.65 ms │ 1776.08 / 1879.19 ±65.89 / 1979.84 ms │     no change │
│ QQuery 73 │        10.21 / 10.56 ±0.21 / 10.77 ms │        10.20 / 12.46 ±3.93 / 20.31 ms │  1.18x slower │
│ QQuery 74 │     180.76 / 182.89 ±1.31 / 184.70 ms │     176.94 / 178.61 ±1.73 / 181.78 ms │     no change │
│ QQuery 75 │     146.50 / 149.01 ±3.87 / 156.72 ms │     149.10 / 152.22 ±5.07 / 162.23 ms │     no change │
│ QQuery 76 │        35.39 / 36.29 ±0.64 / 37.22 ms │        35.69 / 36.10 ±0.31 / 36.54 ms │     no change │
│ QQuery 77 │        60.94 / 61.37 ±0.42 / 62.10 ms │        61.32 / 62.44 ±0.93 / 64.04 ms │     no change │
│ QQuery 78 │     218.70 / 223.95 ±4.69 / 231.37 ms │     206.06 / 211.21 ±7.00 / 225.01 ms │ +1.06x faster │
│ QQuery 79 │        66.73 / 66.98 ±0.20 / 67.23 ms │        66.72 / 67.80 ±0.94 / 69.50 ms │     no change │
│ QQuery 80 │      98.99 / 100.32 ±1.34 / 102.17 ms │      97.80 / 101.55 ±3.23 / 107.55 ms │     no change │
│ QQuery 81 │        26.07 / 26.51 ±0.52 / 27.51 ms │        26.40 / 26.58 ±0.18 / 26.89 ms │     no change │
│ QQuery 82 │        16.05 / 16.37 ±0.23 / 16.70 ms │        16.75 / 16.98 ±0.24 / 17.43 ms │     no change │
│ QQuery 83 │        34.15 / 36.42 ±4.29 / 45.00 ms │        34.23 / 34.48 ±0.18 / 34.72 ms │ +1.06x faster │
│ QQuery 84 │        29.32 / 30.00 ±0.47 / 30.71 ms │        29.37 / 29.59 ±0.22 / 29.96 ms │     no change │
│ QQuery 85 │     102.39 / 103.13 ±0.53 / 103.76 ms │     102.19 / 105.97 ±4.73 / 114.90 ms │     no change │
│ QQuery 86 │        25.11 / 25.39 ±0.23 / 25.73 ms │        25.62 / 26.11 ±0.40 / 26.70 ms │     no change │
│ QQuery 87 │        61.75 / 63.15 ±2.26 / 67.65 ms │        64.96 / 65.73 ±0.61 / 66.41 ms │     no change │
│ QQuery 88 │        63.26 / 63.86 ±0.41 / 64.41 ms │        62.84 / 63.35 ±0.40 / 63.91 ms │     no change │
│ QQuery 89 │        35.39 / 35.73 ±0.37 / 36.35 ms │        36.12 / 38.85 ±4.66 / 48.14 ms │  1.09x slower │
│ QQuery 90 │        16.88 / 17.01 ±0.11 / 17.19 ms │        17.07 / 17.26 ±0.15 / 17.45 ms │     no change │
│ QQuery 91 │        44.55 / 44.89 ±0.26 / 45.32 ms │        45.03 / 45.41 ±0.31 / 45.92 ms │     no change │
│ QQuery 92 │        29.63 / 32.67 ±4.50 / 41.57 ms │        30.57 / 30.96 ±0.29 / 31.42 ms │ +1.06x faster │
│ QQuery 93 │        49.59 / 50.74 ±1.08 / 52.15 ms │        50.21 / 51.83 ±1.05 / 53.21 ms │     no change │
│ QQuery 94 │        38.41 / 38.68 ±0.26 / 39.08 ms │        38.64 / 39.18 ±0.59 / 40.14 ms │     no change │
│ QQuery 95 │        80.01 / 80.61 ±0.53 / 81.47 ms │        80.78 / 85.96 ±6.55 / 98.67 ms │  1.07x slower │
│ QQuery 96 │        23.88 / 25.71 ±2.78 / 31.24 ms │        23.79 / 24.07 ±0.38 / 24.83 ms │ +1.07x faster │
│ QQuery 97 │        52.09 / 53.78 ±1.67 / 56.28 ms │        55.74 / 56.28 ±0.64 / 57.48 ms │     no change │
│ QQuery 98 │        42.65 / 42.91 ±0.32 / 43.52 ms │        43.58 / 45.31 ±2.39 / 50.02 ms │  1.06x slower │
│ QQuery 99 │        69.56 / 69.75 ±0.13 / 69.95 ms │        72.95 / 73.37 ±0.39 / 74.07 ms │  1.05x slower │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                           ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 9557.87ms │
│ Total Time (add-blocks-impl-from-scratch)   │ 9711.02ms │
│ Average Time (HEAD)                         │   96.54ms │
│ Average Time (add-blocks-impl-from-scratch) │   98.09ms │
│ Queries Faster                              │         7 │
│ Queries Slower                              │        15 │
│ Queries with No Change                      │        77 │
│ Queries with Failure                        │         0 │
└─────────────────────────────────────────────┴───────────┘

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	50.0s
Peak memory	2.1 GiB
Avg memory	1.5 GiB
CPU user	209.5s
CPU sys	5.5s
Peak spill	0 B

tpcds — branch

Metric	Value
Wall time	50.0s
Peak memory	2.1 GiB
Avg memory	1.5 GiB
CPU user	206.9s
CPU sys	5.8s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-09-03T22:51:44Z

🤖 Benchmark completed (GKE) | trigger

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

Comparing add-blocks-impl-from-scratch (5711b81) to d2b626c (merge-base) diff

Run configuration

run benchmark clickbench_partitioned

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 add-blocks-impl-from-scratch
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ add-blocks-impl-from-scratch ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │    1.27 ms │                      6.10 ms │  4.82x slower │
│ QQuery 1  │   12.20 ms │                     12.69 ms │     no change │
│ QQuery 2  │   36.98 ms │                     36.90 ms │     no change │
│ QQuery 3  │   31.67 ms │                     32.16 ms │     no change │
│ QQuery 4  │  238.15 ms │                    317.09 ms │  1.33x slower │
│ QQuery 5  │  285.32 ms │                         FAIL │  incomparable │
│ QQuery 6  │    1.29 ms │                      5.35 ms │  4.13x slower │
│ QQuery 7  │   13.96 ms │                     14.13 ms │     no change │
│ QQuery 8  │  337.17 ms │                    379.76 ms │  1.13x slower │
│ QQuery 9  │  457.99 ms │                    557.14 ms │  1.22x slower │
│ QQuery 10 │   72.33 ms │                     73.68 ms │     no change │
│ QQuery 11 │   83.28 ms │                     87.49 ms │  1.05x slower │
│ QQuery 12 │  273.90 ms │                         FAIL │  incomparable │
│ QQuery 13 │  383.00 ms │                         FAIL │  incomparable │
│ QQuery 14 │  291.77 ms │                    316.81 ms │  1.09x slower │
│ QQuery 15 │  279.49 ms │                    375.23 ms │  1.34x slower │
│ QQuery 16 │  631.43 ms │                    724.37 ms │  1.15x slower │
│ QQuery 17 │  635.74 ms │                    729.78 ms │  1.15x slower │
│ QQuery 18 │ 1301.27 ms │                   1400.07 ms │  1.08x slower │
│ QQuery 19 │   27.96 ms │                     28.04 ms │     no change │
│ QQuery 20 │  518.55 ms │                    515.09 ms │     no change │
│ QQuery 21 │  519.97 ms │                    522.23 ms │     no change │
│ QQuery 22 │  995.15 ms │                   1000.98 ms │     no change │
│ QQuery 23 │ 3099.55 ms │                   3083.88 ms │     no change │
│ QQuery 24 │   41.35 ms │                     41.34 ms │     no change │
│ QQuery 25 │  112.80 ms │                    111.28 ms │     no change │
│ QQuery 26 │   42.28 ms │                     41.93 ms │     no change │
│ QQuery 27 │  515.07 ms │                    535.73 ms │     no change │
│ QQuery 28 │ 2936.48 ms │                         FAIL │  incomparable │
│ QQuery 29 │   41.44 ms │                     41.05 ms │     no change │
│ QQuery 30 │  313.21 ms │                    336.65 ms │  1.07x slower │
│ QQuery 31 │  285.80 ms │                    286.74 ms │     no change │
│ QQuery 32 │  950.42 ms │                    997.57 ms │     no change │
│ QQuery 33 │ 1495.87 ms │                         FAIL │  incomparable │
│ QQuery 34 │ 1623.43 ms │                         FAIL │  incomparable │
│ QQuery 35 │  310.64 ms │                    388.99 ms │  1.25x slower │
│ QQuery 36 │   73.43 ms │                         FAIL │  incomparable │
│ QQuery 37 │   37.27 ms │                         FAIL │  incomparable │
│ QQuery 38 │   43.95 ms │                     41.60 ms │ +1.06x faster │
│ QQuery 39 │  154.04 ms │                    154.12 ms │     no change │
│ QQuery 40 │   15.40 ms │                     15.03 ms │     no change │
│ QQuery 41 │   14.85 ms │                     14.50 ms │     no change │
│ QQuery 42 │   14.58 ms │                     14.74 ms │     no change │
└───────────┴────────────┴──────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                           ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 12443.02ms │
│ Total Time (add-blocks-impl-from-scratch)   │ 13240.27ms │
│ Average Time (HEAD)                         │   355.51ms │
│ Average Time (add-blocks-impl-from-scratch) │   378.29ms │
│ Queries Faster                              │          1 │
│ Queries Slower                              │         13 │
│ Queries with No Change                      │         21 │
│ Queries with Failure                        │          8 │
└─────────────────────────────────────────────┴────────────┘

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

Comparing HEAD and add-blocks-impl-from-scratch
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃          add-blocks-impl-from-scratch ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.27 / 4.28 ±5.85 / 15.97 ms │          6.10 / 9.08 ±5.46 / 19.99 ms │  2.12x slower │
│ QQuery 1  │        12.20 / 12.73 ±0.42 / 13.23 ms │        12.69 / 12.84 ±0.11 / 13.01 ms │     no change │
│ QQuery 2  │        36.98 / 37.38 ±0.22 / 37.56 ms │        36.90 / 37.10 ±0.24 / 37.50 ms │     no change │
│ QQuery 3  │        31.67 / 32.43 ±0.68 / 33.60 ms │        32.16 / 32.87 ±0.52 / 33.55 ms │     no change │
│ QQuery 4  │     238.15 / 243.12 ±3.96 / 249.24 ms │     317.09 / 321.73 ±3.30 / 327.34 ms │  1.32x slower │
│ QQuery 5  │     285.32 / 290.63 ±3.24 / 294.39 ms │                                  FAIL │  incomparable │
│ QQuery 6  │           1.29 / 1.46 ±0.24 / 1.93 ms │           5.35 / 6.88 ±0.84 / 7.66 ms │  4.71x slower │
│ QQuery 7  │        13.96 / 14.32 ±0.31 / 14.86 ms │        14.13 / 16.05 ±3.31 / 22.67 ms │  1.12x slower │
│ QQuery 8  │     337.17 / 343.33 ±4.88 / 351.60 ms │     379.76 / 383.65 ±6.17 / 395.95 ms │  1.12x slower │
│ QQuery 9  │     457.99 / 471.88 ±9.33 / 487.09 ms │     557.14 / 569.33 ±6.53 / 576.81 ms │  1.21x slower │
│ QQuery 10 │        72.33 / 73.31 ±1.02 / 74.66 ms │        73.68 / 75.26 ±1.01 / 76.37 ms │     no change │
│ QQuery 11 │        83.28 / 83.59 ±0.18 / 83.81 ms │        87.49 / 88.43 ±0.92 / 90.01 ms │  1.06x slower │
│ QQuery 12 │     273.90 / 279.46 ±4.60 / 286.65 ms │                                  FAIL │  incomparable │
│ QQuery 13 │     383.00 / 394.72 ±9.92 / 409.84 ms │                                  FAIL │  incomparable │
│ QQuery 14 │     291.77 / 298.96 ±5.39 / 306.36 ms │     316.81 / 318.84 ±1.46 / 320.85 ms │  1.07x slower │
│ QQuery 15 │     279.49 / 284.34 ±3.72 / 290.87 ms │     375.23 / 379.84 ±3.94 / 387.10 ms │  1.34x slower │
│ QQuery 16 │     631.43 / 641.86 ±7.44 / 651.80 ms │     724.37 / 733.98 ±6.58 / 742.41 ms │  1.14x slower │
│ QQuery 17 │     635.74 / 646.91 ±9.78 / 662.81 ms │     729.78 / 737.20 ±4.70 / 744.54 ms │  1.14x slower │
│ QQuery 18 │ 1301.27 / 1331.07 ±19.40 / 1357.45 ms │ 1400.07 / 1427.79 ±16.49 / 1445.98 ms │  1.07x slower │
│ QQuery 19 │       27.96 / 37.18 ±18.18 / 73.54 ms │        28.04 / 28.30 ±0.24 / 28.65 ms │ +1.31x faster │
│ QQuery 20 │    518.55 / 533.54 ±16.30 / 563.64 ms │    515.09 / 527.73 ±10.02 / 544.50 ms │     no change │
│ QQuery 21 │     519.97 / 525.97 ±4.18 / 532.17 ms │     522.23 / 528.84 ±4.30 / 534.03 ms │     no change │
│ QQuery 22 │  995.15 / 1010.96 ±16.23 / 1038.12 ms │ 1000.98 / 1015.44 ±17.51 / 1038.98 ms │     no change │
│ QQuery 23 │ 3099.55 / 3134.80 ±25.92 / 3167.42 ms │ 3083.88 / 3132.08 ±31.39 / 3166.37 ms │     no change │
│ QQuery 24 │        41.35 / 45.01 ±4.07 / 51.61 ms │       41.34 / 49.82 ±16.27 / 82.36 ms │  1.11x slower │
│ QQuery 25 │     112.80 / 113.80 ±0.82 / 114.84 ms │     111.28 / 112.87 ±1.95 / 116.48 ms │     no change │
│ QQuery 26 │        42.28 / 43.19 ±1.33 / 45.71 ms │        41.93 / 43.33 ±1.96 / 47.14 ms │     no change │
│ QQuery 27 │     515.07 / 523.66 ±7.42 / 536.92 ms │     535.73 / 540.74 ±4.89 / 548.96 ms │     no change │
│ QQuery 28 │ 2936.48 / 2956.96 ±17.61 / 2988.85 ms │                                  FAIL │  incomparable │
│ QQuery 29 │        41.44 / 50.46 ±7.75 / 59.43 ms │        41.05 / 44.92 ±4.33 / 52.53 ms │ +1.12x faster │
│ QQuery 30 │     313.21 / 320.45 ±7.85 / 334.32 ms │     336.65 / 344.13 ±5.44 / 351.29 ms │  1.07x slower │
│ QQuery 31 │    285.80 / 298.07 ±10.40 / 310.77 ms │     286.74 / 293.83 ±4.26 / 298.38 ms │     no change │
│ QQuery 32 │    950.42 / 968.30 ±13.16 / 990.52 ms │  997.57 / 1033.02 ±38.98 / 1106.70 ms │  1.07x slower │
│ QQuery 33 │ 1495.87 / 1558.70 ±45.48 / 1618.30 ms │                                  FAIL │  incomparable │
│ QQuery 34 │ 1623.43 / 1645.41 ±15.64 / 1668.07 ms │                                  FAIL │  incomparable │
│ QQuery 35 │    310.64 / 366.84 ±72.58 / 506.63 ms │     388.99 / 392.71 ±2.86 / 397.46 ms │  1.07x slower │
│ QQuery 36 │        73.43 / 85.58 ±6.77 / 92.53 ms │                                  FAIL │  incomparable │
│ QQuery 37 │        37.27 / 39.88 ±3.19 / 46.15 ms │                                  FAIL │  incomparable │
│ QQuery 38 │        43.95 / 45.58 ±0.93 / 46.84 ms │        41.60 / 47.37 ±5.03 / 54.67 ms │     no change │
│ QQuery 39 │     154.04 / 163.56 ±7.35 / 176.59 ms │     154.12 / 164.29 ±8.27 / 178.91 ms │     no change │
│ QQuery 40 │        15.40 / 15.92 ±0.74 / 17.36 ms │        15.03 / 17.33 ±3.73 / 24.76 ms │  1.09x slower │
│ QQuery 41 │        14.85 / 16.42 ±2.78 / 21.98 ms │        14.50 / 14.73 ±0.21 / 15.09 ms │ +1.12x faster │
│ QQuery 42 │        14.58 / 14.70 ±0.08 / 14.79 ms │        14.74 / 17.36 ±4.78 / 26.91 ms │  1.18x slower │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                           ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 12749.38ms │
│ Total Time (add-blocks-impl-from-scratch)   │ 13499.72ms │
│ Average Time (HEAD)                         │   364.27ms │
│ Average Time (add-blocks-impl-from-scratch) │   385.71ms │
│ Queries Faster                              │          3 │
│ Queries Slower                              │         18 │
│ Queries with No Change                      │         14 │
│ Queries with Failure                        │          8 │
└─────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	105.0s
Peak memory	11.5 GiB
Avg memory	4.5 GiB
CPU user	1021.8s
CPU sys	74.8s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	70.0s
Peak memory	11.5 GiB
Avg memory	3.7 GiB
CPU user	691.2s
CPU sys	48.3s
Peak spill	0 B

File an issue against this benchmark runner

rluvaton added 30 commits August 31, 2026 10:54

copy paste aggregates as is to aggregates_blocked

5cdf7d5

update imports and make it compile

cc0e79a

add steps

5979c90

use existing structs from aggregate to avoid full duplication

16e3848

rename AggregateExec in blocked to BlockedAggregateExec

1a2e531

update AggregateExec::try_new to return the BlockedAggregateExec …

f0f7b8d

…so all usages will be migrated

add fallback to BlockedAggregateExec so unsupported logic will be for…

672ebb5

…warded

Rename GroupValues to BlockedGroupValues and GroupColumn to `Bl…

41b926c

…ockedGroupColumn`

Add BlockedIndex and BlockedEmitTo

358d618

Change in group indices to be stored internally as BlocksIndex and …

464f8fe

…not `usize`

Remove all BlockedGroupValues impl and replace with adapter

e5d3895

Add batch_size to BlockedGroupValues for later emit change

1e9c6d1

fallback to non blocked in grouped_hash_stream and removed duplicat…

2ae5d2c

…e grouped_hash_stream code

add blocked emit to

6970f8c

changed intern to get BlocksIndex and not usize

a605809

create blocked groups accumulator

e6cdab2

remove print

d6a5479

fix compile for tests

50796d5

add group indices to ordering

f76ba2a

Merge branch 'main' into add-blocks-impl-from-scratch

22e0f32

# Conflicts: # datafusion/expr-common/src/groups_accumulator.rs

fix emit too much

9082071

add blocked group by boolean

0c67d68

add state_preserving to blocked group accumulator adapter

923ee62

add evaluate_preserving to blocked group accumulator adapter

0aeb359

fix compile for BlockedGroupsSelection

7fb5b7a

impl blocked group accumulator for sum

f511677

fix group by boolean

11216d2

add support for blocked min max

ee5f265

add more blocked helpers and tests

98e7c1a

re-export and fix clippy

e34750f

rluvaton added 2 commits September 3, 2026 19:40

add support for group by single bytes column

89e615f

impl bytes view

d9d288b

github-actions Bot added logical-expr Logical plan and expressions physical-expr Changes to the physical-expr crates core Core DataFusion crate common Related to common crate functions Changes to functions implementation physical-plan Changes to the physical-plan crate labels Sep 3, 2026

rluvaton changed the title ~~copy paste aggregates as is to aggregates_blocked~~ aggregate blocked impl Sep 3, 2026

github-actions Bot added the auto detected api change Auto detected API change label Sep 3, 2026

rluvaton added 5 commits September 3, 2026 23:31

impl multi group by with boolean column support

ec2a018

impl multi group by with primitive column support

f6e9ffe

fix byte view map

9b64a45

impl multi group by with bytes column support

4b685db

impl multi group by with bytes view column support

5711b81

Uh oh!

Conversation

rluvaton commented Sep 3, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Which issue does this PR close?

Rationale for this change

What changes are included in this PR?

What is the testing strategy for this PR?

Are there any user-facing changes?

Uh oh!

rluvaton commented Sep 3, 2026

Uh oh!

adriangbot commented Sep 3, 2026

Uh oh!

adriangbot commented Sep 3, 2026

Uh oh!

adriangbot commented Sep 3, 2026

Uh oh!

github-actions Bot commented Sep 3, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

adriangbot commented Sep 3, 2026

Uh oh!

adriangbot commented Sep 3, 2026

Uh oh!

adriangbot commented Sep 3, 2026

Uh oh!

rluvaton commented Sep 3, 2026

Uh oh!

adriangbot commented Sep 3, 2026

Uh oh!

adriangbot commented Sep 3, 2026

Uh oh!

adriangbot commented Sep 3, 2026

Uh oh!

adriangbot commented Sep 3, 2026

Uh oh!

adriangbot commented Sep 3, 2026

Uh oh!

adriangbot commented Sep 3, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

rluvaton commented Sep 3, 2026 •

edited

Loading

github-actions Bot commented Sep 3, 2026 •

edited

Loading