perf: bulk-append contiguous buffered runs in sort merge join#23673
Open
andygrove wants to merge 1 commit into
Open
perf: bulk-append contiguous buffered runs in sort merge join#23673andygrove wants to merge 1 commit into
andygrove wants to merge 1 commit into
Conversation
The materializing sort-merge-join stream advanced one output row at a time in join_partial. When a streamed row matches a run of buffered rows (high fan-out), each matched row cost a VecDeque index lookup in scanning_idx plus two more in scanning_advance, along with two individual builder appends. The matched buffered rows within a batch form a contiguous index range, so append the whole run at once via append_output_range (append_value_n + append_slice) and advance the scan cursor by the run length. A len == 1 fast path keeps the common one-row-per-key case free of scratch-buffer overhead. Benchmarks (sort_merge_join): inner join with 1:10 fan-out improves ~23% (13.28ms -> 10.18ms); 1:1 cases are neutral.
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #23673 +/- ##
=======================================
Coverage 80.67% 80.67%
=======================================
Files 1086 1086
Lines 366800 366853 +53
Branches 366800 366853 +53
=======================================
+ Hits 295912 295962 +50
+ Misses 53260 53256 -4
- Partials 17628 17635 +7 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
Member
Author
|
run benchmark sort_merge_join |
|
Hi @andygrove, thanks for the request (#23673 (comment)). Only whitelisted users can trigger benchmarks. Allowed users: Dandandan, Fokko, Jefffrey, Omega359, Rachelint, adriangb, alamb, asubiotto, brunal, buraksenn, cetra3, codephage2020, coderfender, comphead, erenavsarogullari, etseidl, friendlymatthew, gabotechs, geoffreyclaude, grtlr, haohuaijin, jonathanc-n, kevinjqliu, klion26, kosiew, kumarUjjawal, kunalsinghdadhwal, liamzwbao, mbutrovich, mkleen, mzabaluev, neilconway, rluvaton, sdf-jkl, timsaucer, xudong963, zhuqi-lucas. File an issue against this benchmark runner |
Member
Author
|
@mbutrovich @comphead could you trigger the benchmark run for me? I'm not on the list. |
Contributor
|
run benchmark sort_merge_join,tpch,tpcds |
|
Hi @comphead, thanks for the request (#23673 (comment)). Usage: Any benchmark name is accepted: Per-side configuration ( env:
# shared env is inherited by BOTH the build and the run, so build
# flags go here. Builds default to no debuginfo for speed; opt back
# in for hung-job gdb dumps and cap jobs to stay within memory:
CARGO_PROFILE_RELEASE_DEBUG: "1"
CARGO_BUILD_JOBS: "1"
baseline:
ref: v45.0.0
env:
# per-side env only reaches the benchmark run, not the build
DATAFUSION_RUNTIME_MEMORY_LIMIT: 1G
changed:
ref: v46.0.0
env:
DATAFUSION_RUNTIME_MEMORY_LIMIT: 2GFile an issue against this benchmark runner |
Contributor
|
run benchmark sort_merge_join tpch tpcds |
|
🤖 Benchmark running (GKE) | trigger CPU Details (lscpu)Comparing smj-bulk-range-append (a7fa061) to 1e58928 (merge-base) diff using: sort_merge_join File an issue against this benchmark runner |
|
🤖 Benchmark running (GKE) | trigger CPU Details (lscpu)Comparing smj-bulk-range-append (a7fa061) to 1e58928 (merge-base) diff using: tpch File an issue against this benchmark runner |
|
🤖 Benchmark running (GKE) | trigger CPU Details (lscpu)Comparing smj-bulk-range-append (a7fa061) to 1e58928 (merge-base) diff using: tpcds File an issue against this benchmark runner |
|
Benchmark for this request failed. Last 20 lines of output: Click to expandFile an issue against this benchmark runner |
Contributor
|
run benchmark tpch tpcds env: |
|
🤖 Benchmark running (GKE) | trigger CPU Details (lscpu)Comparing smj-bulk-range-append (a7fa061) to 1e58928 (merge-base) diff using: tpch File an issue against this benchmark runner |
|
🤖 Benchmark running (GKE) | trigger CPU Details (lscpu)Comparing smj-bulk-range-append (a7fa061) to 1e58928 (merge-base) diff using: tpcds File an issue against this benchmark runner |
|
🤖 Benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usagetpch — base (merge-base)
tpch — branch
File an issue against this benchmark runner |
|
🤖 Benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usagetpch — base (merge-base)
tpch — branch
File an issue against this benchmark runner |
|
🤖 Benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usagetpcds — base (merge-base)
tpcds — branch
File an issue against this benchmark runner |
|
🤖 Benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usagetpcds — base (merge-base)
tpcds — branch
File an issue against this benchmark runner |
Member
Author
|
hmmm... benchmark results seem pretty mixed so far |
Contributor
|
run benchmark tpch10 |
|
🤖 Benchmark running (GKE) | trigger CPU Details (lscpu)Comparing smj-bulk-range-append (a7fa061) to 1e58928 (merge-base) diff using: tpch10 File an issue against this benchmark runner |
|
🤖 Benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usagetpch10 — base (merge-base)
tpch10 — branch
File an issue against this benchmark runner |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Which issue does this PR close?
Rationale for this change
The
MaterializingSortMergeJoinStream(used for inner/left/right/full joins) produces output one row at a time injoin_partial. When a single streamed row matches a run of buffered rows — i.e. a high fan-out join — each matched row incurs:VecDequeindex lookup inscanning_idx(),VecDequeindex computations insidescanning_advance(), andUInt64Builder::append_valuecalls.The buffered rows that match within a single buffered batch always form a contiguous index range, so this per-row work is unnecessary overhead that scales with the number of output rows.
What changes are included in this PR?
StreamedBatch::append_output_range, which appends a contiguous run of buffered indices against the current streamed index in bulk (append_value_n+append_slice) instead of one pair at a time. A reusable scratch buffer (buffered_index_scratch) materializes the index range, and alen == 1fast path keeps the common one-row-per-key case free of scratch-buffer overhead.BufferedData::scanning_advance_by(n)andscanning_batch_remaining()so the scan cursor can advance by a whole run at once.join_partialnow computes the run length (capped by the remaining output-batch capacity) and appends the whole run in a single step. The FULL-join null-joined branch is likewise updated toextenda range rather than pushing per row.No behavior change — output is identical, only the way indices are accumulated changes.
Benchmark results (
cargo bench --bench sort_merge_join)inner_1to10(1:10 fan-out, 1M output rows)inner_1to1left_1to1_unmatchedThe semi/anti benchmarks use a different stream (
BitwiseSortMergeJoinStream) that is unchanged.Are these changes tested?
Covered by existing tests:
datafusion-physical-planpass.sort_merge_join.sltsqllogictest passes.Are there any user-facing changes?
No.