You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@datafusion.apache.org by "Rachelint (via GitHub)" <gi...@apache.org> on 2026/10/07 05:07:58 UTC
[PR] bench: isolate continuous partial flushing against main [datafusion]
Rachelint opened a new pull request, #26098:
URL: https://github.com/apache/datafusion/pull/26098
## Which issue does this PR close?
None. Benchmark-only experiment based on #26097 (`85b64a74cec1eab1e5a590234cd242fa0ece843d`).
## Rationale for this change
Compare continuous Partial flushing with ordinary Final aggregation against main, and expose independent switches for follow-up experiments.
## What changes are included in this PR?
- Revert the fixed-width-only eligibility change (`88d36d4d`) so string keys and string state are eligible again; preserve the original nested-state exclusions.
- Preserve #26097's 2 MiB allocated-byte Partial threshold, in addition to the group-count threshold.
- Add three experimental execution options, all defaulting to true: `hash_aggregate_bucket_compaction`, `hash_aggregate_detect_repeated_groups`, and `hash_aggregate_final_buckets`.
- Disabling recurrence detection bypasses emitted-key hashing/sampling and keeps threshold flushing active. Disabling Final buckets leaves Partial flushing enabled.
- No StringView allocation reuse, payload reserve, or column-wise scatter changes from #25959 are included.
Requested comparison: all three switches false, `hash_aggregate_bucket_threshold=262144`, against main at `7f0ab426eb84a967a93c42ee0bd6e928e49adeb6`. Existing skip-partial aggregation remains enabled. Main uses its ordinary behavior. This compares the full experiment with main, not just the incremental switch diff.
## What is the testing strategy for this PR?
Added SQLLogicTest coverage for integer, long StringView, and mixed keys with all three switches disabled, plus Final buckets enabled with compaction disabled. Updated configuration documentation and information_schema fixtures.
- `cargo fmt --all` and format check: passed.
- `cargo clippy --all-targets --all-features -- -D warnings`: passed.
- Configuration docs regenerated; `./ci/scripts/doc_prettier_check.sh --write --allow-dirty`: passed.
- Required workspace extended tests: passed (12,382 Rust tests and 527 SQLLogicTest files).
- ECS runtime source hashes match the three modified Rust source files.
ECS comparison: Rust 1.99.0, `release-nonlto`, 12 partitions on CPUs 0-11, batch size 8192, 99,997,497 partitioned input rows. Two reverse-order rounds; four iterations per cell, first excluded (six warm samples per variant). Unmodified benchmark runner, entire-query wall-time medians:
| Query | main (ms) | Candidate (ms) | Change |
|---|---:|---:|---:|
| Q10 | 106.80 | 110.82 | +3.76% |
| Q11 | 123.25 | 127.66 | +3.58% |
| Q12 | 404.42 | 390.75 | -3.38% |
| Q31 | 460.88 | 456.69 | -0.91% |
| Q32 | 1776.68 | 1790.72 | +0.79% |
| Q33 | 2286.17 | 2016.63 | -11.79% |
| Q34 | 2277.13 | 2048.87 | -10.02% |
| Q35 | 419.98 | 354.63 | -15.56% |
Q33/Q34/Q35 gains repeat in both orders. Q10/Q11 regress slightly; Q31/Q32 remain near flat. The benchmark checks success and output row counts; SQLLogicTest provides value-level correctness coverage. These results compare the full experiment against the pinned current main; they do not isolate unrelated changes between the parent PR base and main.
## Are there any user-facing changes?
Three experimental configuration switches. The parent feature is disabled by default (`hash_aggregate_bucket_threshold=0`). This PR is an experimental benchmark vehicle, not intended for merge.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] bench: isolate continuous partial flushing against main [datafusion]
Posted by "Rachelint (via GitHub)" <gi...@apache.org>.
Rachelint commented on PR #26098:
URL: https://github.com/apache/datafusion/pull/26098#issuecomment-6031699653
run benchmark clickbench_partitioned
baseline:
ref: 7f0ab426eb84a967a93c42ee0bd6e928e49adeb6
env:
DATAFUSION_EXECUTION_HASH_AGGREGATE_BUCKET_THRESHOLD: "262144"
DATAFUSION_EXECUTION_HASH_AGGREGATE_BUCKET_COMPACTION: "false"
DATAFUSION_EXECUTION_HASH_AGGREGATE_DETECT_REPEATED_GROUPS: "false"
DATAFUSION_EXECUTION_HASH_AGGREGATE_FINAL_BUCKETS: "true"
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] bench: isolate continuous partial flushing against main [datafusion]
Posted by "github-actions[bot] (via GitHub)" <gi...@apache.org>.
github-actions[bot] commented on PR #26098:
URL: https://github.com/apache/datafusion/pull/26098#issuecomment-6031435751
<!-- semver-check-comment -->
Thank you for opening this pull request!
Reviewer note: [cargo-semver-checks](https://github.com/obi1kenobi/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>
<summary>Details</summary>
```
Cloning apache/main
Building datafusion-common v55.1.0 (current)
Built [ 32.958s] (current)
Parsing datafusion-common v55.1.0 (current)
Parsed [ 0.064s] (current)
Building datafusion-common v55.1.0 (baseline)
Built [ 31.591s] (baseline)
Parsing datafusion-common v55.1.0 (baseline)
Parsed [ 0.061s] (baseline)
Checking datafusion-common v55.1.0 -> v55.1.0 (no change; assume patch)
Checked [ 1.081s] 223 checks: 222 pass, 1 fail, 0 warn, 31 skip
--- failure constructible_struct_adds_field: struct exhaustively constructible through public API adds field ---
Description:
A pub struct that could be exhaustively constructed with a literal using only public API has a new pub field, breaking existing exhaustive literals.
ref: https://doc.rust-lang.org/reference/expressions/struct-expr.html
impl: https://github.com/obi1kenobi/cargo-semver-checks/tree/v0.50.0/src/lints/constructible_struct_adds_field.ron
Failed in:
field ExecutionOptions.hash_aggregate_bucket_threshold in /home/runner/work/datafusion/datafusion/datafusion/common/src/config.rs:897
field ExecutionOptions.hash_aggregate_bucket_compaction in /home/runner/work/datafusion/datafusion/datafusion/common/src/config.rs:897
field ExecutionOptions.hash_aggregate_detect_repeated_groups in /home/runner/work/datafusion/datafusion/datafusion/common/src/config.rs:897
field ExecutionOptions.hash_aggregate_final_buckets in /home/runner/work/datafusion/datafusion/datafusion/common/src/config.rs:897
Summary semver requires new major version: 1 major and 0 minor checks failed
Finished [ 68.052s] datafusion-common
Building datafusion-physical-plan v55.1.0 (current)
Built [ 36.295s] (current)
Parsing datafusion-physical-plan v55.1.0 (current)
Parsed [ 0.183s] (current)
Building datafusion-physical-plan v55.1.0 (baseline)
Built [ 36.826s] (baseline)
Parsing datafusion-physical-plan v55.1.0 (baseline)
Parsed [ 0.182s] (baseline)
Checking datafusion-physical-plan v55.1.0 -> v55.1.0 (no change; assume patch)
Checked [ 0.995s] 223 checks: 223 pass, 31 skip
Summary no semver update required
Finished [ 76.511s] datafusion-physical-plan
Building datafusion-sqllogictest v55.1.0 (current)
Built [ 90.597s] (current)
Parsing datafusion-sqllogictest v55.1.0 (current)
Parsed [ 0.015s] (current)
Building datafusion-sqllogictest v55.1.0 (baseline)
Built [ 91.322s] (baseline)
Parsing datafusion-sqllogictest v55.1.0 (baseline)
Parsed [ 0.016s] (baseline)
Checking datafusion-sqllogictest v55.1.0 -> v55.1.0 (no change; assume patch)
Checked [ 0.146s] 223 checks: 223 pass, 31 skip
Summary no semver update required
Finished [ 184.804s] datafusion-sqllogictest
```
</details>
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] bench: isolate continuous partial flushing against main [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #26098:
URL: https://github.com/apache/datafusion/pull/26098#issuecomment-6031746642
π€ Benchmark completed (GKE) | [trigger](https://github.com/apache/datafusion/pull/26098#issuecomment-6031351170)
**Instance:** `c4a-highcpu-48` (12 vCPU / 65 GiB)
Comparing codex/partial-flush-controls (fe0866a91902f5a835be6a68713a78c0855c56b1) to 7f0ab426eb84a967a93c42ee0bd6e928e49adeb6 [diff](https://github.com/apache/datafusion/compare/7f0ab426eb84a967a93c42ee0bd6e928e49adeb6..fe0866a91902f5a835be6a68713a78c0855c56b1)
<details><summary>Run configuration</summary>
```yaml
run benchmark clickbench_partitioned
env:
DATAFUSION_EXECUTION_HASH_AGGREGATE_BUCKET_COMPACTION: "false"
DATAFUSION_EXECUTION_HASH_AGGREGATE_BUCKET_THRESHOLD: "262144"
DATAFUSION_EXECUTION_HASH_AGGREGATE_DETECT_REPEATED_GROUPS: "false"
DATAFUSION_EXECUTION_HASH_AGGREGATE_FINAL_BUCKETS: "false"
baseline:
ref: "7f0ab426eb84a967a93c42ee0bd6e928e49adeb6"
```
</details>
<details><summary>CPU Details (lscpu)</summary>
```
Architecture: aarch64
CPU op-mode(s): 64-bit
Byte Order: Little Endian
CPU(s): 48
On-line CPU(s) list: 0-47
Vendor ID: ARM
Model name: Neoverse-V2
Model: 1
Thread(s) per core: 1
Core(s) per cluster: 48
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: 3 MiB (48 instances)
L1i cache: 3 MiB (48 instances)
L2 cache: 96 MiB (48 instances)
L3 cache: 80 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0-47
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>
<details><summary>Details</summary>
<p>
```
Comparing HEAD and codex_partial-flush-controls
--------------------
Benchmark clickbench_partitioned.json
--------------------
βββββββββββββ³βββββββββββββ³βββββββββββββββββββββββββββββββ³ββββββββββββββββ
β Query β HEAD β codex_partial-flush-controls β Change β
β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
β QQuery 0 β 1.21 ms β 1.24 ms β no change β
β QQuery 1 β 11.58 ms β 11.64 ms β no change β
β QQuery 2 β 36.42 ms β 36.26 ms β no change β
β QQuery 3 β 30.61 ms β 30.87 ms β no change β
β QQuery 4 β 230.44 ms β 218.76 ms β +1.05x faster β
β QQuery 5 β 270.55 ms β 272.70 ms β no change β
β QQuery 6 β 1.27 ms β 1.28 ms β no change β
β QQuery 7 β 12.58 ms β 12.64 ms β no change β
β QQuery 8 β 324.38 ms β 314.48 ms β no change β
β QQuery 9 β 458.71 ms β 464.69 ms β no change β
β QQuery 10 β 62.82 ms β 63.66 ms β no change β
β QQuery 11 β 73.18 ms β 75.00 ms β no change β
β QQuery 12 β 255.78 ms β 259.35 ms β no change β
β QQuery 13 β 353.64 ms β 354.30 ms β no change β
β QQuery 14 β 278.07 ms β 269.41 ms β no change β
β QQuery 15 β 275.78 ms β 253.99 ms β +1.09x faster β
β QQuery 16 β 612.06 ms β 567.58 ms β +1.08x faster β
β QQuery 17 β 616.25 ms β 568.66 ms β +1.08x faster β
β QQuery 18 β 1256.26 ms β 1060.54 ms β +1.18x faster β
β QQuery 19 β 26.79 ms β 26.78 ms β no change β
β QQuery 20 β 514.17 ms β 513.87 ms β no change β
β QQuery 21 β 505.37 ms β 504.38 ms β no change β
β QQuery 22 β 993.89 ms β 989.67 ms β no change β
β QQuery 23 β 3113.99 ms β 3034.33 ms β no change β
β QQuery 24 β 39.84 ms β 40.00 ms β no change β
β QQuery 25 β 104.34 ms β 103.28 ms β no change β
β QQuery 26 β 40.29 ms β 40.81 ms β no change β
β QQuery 27 β 505.17 ms β 508.66 ms β no change β
β QQuery 28 β 2865.29 ms β 2854.09 ms β no change β
β QQuery 29 β 38.69 ms β 38.76 ms β no change β
β QQuery 30 β 302.34 ms β 286.32 ms β +1.06x faster β
β QQuery 31 β 282.35 ms β 280.04 ms β no change β
β QQuery 32 β 1111.65 ms β 1077.42 ms β no change β
β QQuery 33 β 1538.73 ms β 1382.94 ms β +1.11x faster β
β QQuery 34 β 1558.35 ms β 1380.60 ms β +1.13x faster β
β QQuery 35 β 307.92 ms β 252.11 ms β +1.22x faster β
β QQuery 36 β 70.05 ms β 62.71 ms β +1.12x faster β
β QQuery 37 β 35.46 ms β 34.42 ms β no change β
β QQuery 38 β 42.30 ms β 39.98 ms β +1.06x faster β
β QQuery 39 β 156.16 ms β 138.25 ms β +1.13x faster β
β QQuery 40 β 13.80 ms β 13.78 ms β no change β
β QQuery 41 β 13.25 ms β 12.95 ms β no change β
β QQuery 42 β 12.90 ms β 12.91 ms β no change β
βββββββββββββ΄βββββββββββββ΄βββββββββββββββββββββββββββββββ΄ββββββββββββββββ
βββββββββββββββββββββββββββββββββββββββββββββββ³βββββββββββββ
β Benchmark Summary β β
β‘βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
β Total Time (HEAD) β 19354.69ms β
β Total Time (codex_partial-flush-controls) β 18466.10ms β
β Average Time (HEAD) β 450.11ms β
β Average Time (codex_partial-flush-controls) β 429.44ms β
β Queries Faster β 12 β
β Queries Slower β 0 β
β Queries with No Change β 31 β
β Queries with Failure β 0 β
βββββββββββββββββββββββββββββββββββββββββββββββ΄βββββββββββββ
Distribution per query (min / mean Β±stddev / max):
Comparing HEAD and codex_partial-flush-controls
--------------------
Benchmark clickbench_partitioned.json
--------------------
βββββββββββββ³βββββββββββββββββββββββββββββββββββββββββ³βββββββββββββββββββββββββββββββββββββββββ³ββββββββββββββββ
β Query β HEAD β codex_partial-flush-controls β Change β
β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
β QQuery 0 β 1.21 / 3.91 Β±5.34 / 14.58 ms β 1.24 / 4.01 Β±5.47 / 14.94 ms β no change β
β QQuery 1 β 11.58 / 11.68 Β±0.08 / 11.77 ms β 11.64 / 11.73 Β±0.09 / 11.88 ms β no change β
β QQuery 2 β 36.42 / 36.73 Β±0.28 / 37.10 ms β 36.26 / 36.57 Β±0.16 / 36.69 ms β no change β
β QQuery 3 β 30.61 / 30.99 Β±0.56 / 32.10 ms β 30.87 / 31.07 Β±0.13 / 31.21 ms β no change β
β QQuery 4 β 230.44 / 233.59 Β±2.41 / 236.71 ms β 218.76 / 223.81 Β±3.48 / 228.94 ms β no change β
β QQuery 5 β 270.55 / 273.31 Β±1.76 / 276.11 ms β 272.70 / 279.57 Β±6.52 / 291.80 ms β no change β
β QQuery 6 β 1.27 / 1.40 Β±0.20 / 1.79 ms β 1.28 / 1.42 Β±0.23 / 1.87 ms β no change β
β QQuery 7 β 12.58 / 12.76 Β±0.13 / 12.97 ms β 12.64 / 12.77 Β±0.08 / 12.87 ms β no change β
β QQuery 8 β 324.38 / 329.99 Β±3.25 / 334.23 ms β 314.48 / 317.98 Β±4.07 / 325.78 ms β no change β
β QQuery 9 β 458.71 / 474.42 Β±11.61 / 490.28 ms β 464.69 / 469.09 Β±4.27 / 476.68 ms β no change β
β QQuery 10 β 62.82 / 63.46 Β±0.64 / 64.29 ms β 63.66 / 68.99 Β±8.49 / 85.92 ms β 1.09x slower β
β QQuery 11 β 73.18 / 76.70 Β±5.54 / 87.70 ms β 75.00 / 76.80 Β±1.56 / 78.69 ms β no change β
β QQuery 12 β 255.78 / 261.97 Β±3.50 / 266.26 ms β 259.35 / 261.13 Β±1.47 / 262.93 ms β no change β
β QQuery 13 β 353.64 / 362.46 Β±6.78 / 373.92 ms β 354.30 / 358.59 Β±3.26 / 363.97 ms β no change β
β QQuery 14 β 278.07 / 286.50 Β±14.02 / 314.43 ms β 269.41 / 281.46 Β±8.66 / 291.11 ms β no change β
β QQuery 15 β 275.78 / 284.90 Β±5.25 / 290.99 ms β 253.99 / 254.86 Β±0.80 / 256.00 ms β +1.12x faster β
β QQuery 16 β 612.06 / 621.51 Β±9.43 / 639.51 ms β 567.58 / 574.81 Β±9.47 / 593.29 ms β +1.08x faster β
β QQuery 17 β 616.25 / 625.09 Β±7.05 / 635.21 ms β 568.66 / 576.18 Β±6.06 / 583.05 ms β +1.08x faster β
β QQuery 18 β 1256.26 / 1269.35 Β±9.01 / 1280.07 ms β 1060.54 / 1071.11 Β±8.09 / 1084.80 ms β +1.19x faster β
β QQuery 19 β 26.79 / 28.91 Β±3.74 / 36.38 ms β 26.78 / 27.03 Β±0.18 / 27.32 ms β +1.07x faster β
β QQuery 20 β 514.17 / 522.33 Β±12.11 / 546.31 ms β 513.87 / 526.04 Β±15.88 / 556.47 ms β no change β
β QQuery 21 β 505.37 / 524.21 Β±14.48 / 550.17 ms β 504.38 / 508.47 Β±3.91 / 514.72 ms β no change β
β QQuery 22 β 993.89 / 1041.69 Β±45.01 / 1117.23 ms β 989.67 / 992.26 Β±1.49 / 993.73 ms β no change β
β QQuery 23 β 3113.99 / 3207.69 Β±121.59 / 3445.15 ms β 3034.33 / 3256.28 Β±188.28 / 3508.41 ms β no change β
β QQuery 24 β 39.84 / 66.21 Β±51.92 / 170.05 ms β 40.00 / 43.05 Β±4.23 / 51.34 ms β +1.54x faster β
β QQuery 25 β 104.34 / 186.95 Β±98.24 / 309.61 ms β 103.28 / 193.89 Β±109.84 / 357.04 ms β no change β
β QQuery 26 β 40.29 / 41.30 Β±1.68 / 44.65 ms β 40.81 / 41.67 Β±0.82 / 43.16 ms β no change β
β QQuery 27 β 505.17 / 513.99 Β±9.24 / 530.57 ms β 508.66 / 517.40 Β±6.20 / 527.81 ms β no change β
β QQuery 28 β 2865.29 / 2887.14 Β±29.68 / 2944.92 ms β 2854.09 / 2875.05 Β±13.17 / 2891.95 ms β no change β
β QQuery 29 β 38.69 / 39.03 Β±0.27 / 39.38 ms β 38.76 / 39.15 Β±0.35 / 39.77 ms β no change β
β QQuery 30 β 302.34 / 308.55 Β±5.01 / 317.45 ms β 286.32 / 295.37 Β±9.45 / 312.64 ms β no change β
β QQuery 31 β 282.35 / 299.98 Β±24.10 / 347.66 ms β 280.04 / 293.57 Β±10.16 / 307.21 ms β no change β
β QQuery 32 β 1111.65 / 1225.20 Β±161.35 / 1542.68 ms β 1077.42 / 1249.77 Β±118.54 / 1447.87 ms β no change β
β QQuery 33 β 1538.73 / 1593.76 Β±29.14 / 1625.32 ms β 1382.94 / 1406.74 Β±17.82 / 1424.04 ms β +1.13x faster β
β QQuery 34 β 1558.35 / 1682.00 Β±180.45 / 2038.75 ms β 1380.60 / 1423.30 Β±48.02 / 1516.62 ms β +1.18x faster β
β QQuery 35 β 307.92 / 352.35 Β±72.46 / 497.00 ms β 252.11 / 271.55 Β±30.90 / 332.54 ms β +1.30x faster β
β QQuery 36 β 70.05 / 77.59 Β±8.77 / 94.18 ms β 62.71 / 67.30 Β±3.53 / 71.65 ms β +1.15x faster β
β QQuery 37 β 35.46 / 39.04 Β±3.99 / 46.06 ms β 34.42 / 34.86 Β±0.47 / 35.75 ms β +1.12x faster β
β QQuery 38 β 42.30 / 43.62 Β±1.40 / 45.75 ms β 39.98 / 44.20 Β±4.26 / 52.35 ms β no change β
β QQuery 39 β 156.16 / 160.12 Β±3.26 / 165.42 ms β 138.25 / 143.02 Β±4.08 / 148.25 ms β +1.12x faster β
β QQuery 40 β 13.80 / 14.18 Β±0.31 / 14.57 ms β 13.78 / 13.95 Β±0.21 / 14.34 ms β no change β
β QQuery 41 β 13.25 / 13.40 Β±0.19 / 13.77 ms β 12.95 / 13.40 Β±0.36 / 14.05 ms β no change β
β QQuery 42 β 12.90 / 13.88 Β±1.70 / 17.28 ms β 12.91 / 15.34 Β±4.37 / 24.07 ms β 1.11x slower β
βββββββββββββ΄βββββββββββββββββββββββββββββββββββββββββ΄βββββββββββββββββββββββββββββββββββββββββ΄ββββββββββββββββ
βββββββββββββββββββββββββββββββββββββββββββββββ³βββββββββββββ
β Benchmark Summary β β
β‘βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
β Total Time (HEAD) β 20143.88ms β
β Total Time (codex_partial-flush-controls) β 19204.61ms β
β Average Time (HEAD) β 468.46ms β
β Average Time (codex_partial-flush-controls) β 446.62ms β
β Queries Faster β 12 β
β Queries Slower β 2 β
β Queries with No Change β 29 β
β Queries with Failure β 0 β
βββββββββββββββββββββββββββββββββββββββββββββββ΄βββββββββββββ
```
</p>
</details>
<details><summary>Resource Usage</summary>
**clickbench_partitioned β base (merge-base)**
| Metric | Value |
|--------|-------|
| Wall time | 105.0s |
| Peak memory | 18.3 GiB |
| Avg memory | 5.8 GiB |
| CPU user | 990.1s |
| CPU sys | 94.2s |
| Peak spill | 0 B |
**clickbench_partitioned β branch**
| Metric | Value |
|--------|-------|
| Wall time | 100.0s |
| Peak memory | 18.4 GiB |
| Avg memory | 5.4 GiB |
| CPU user | 947.2s |
| CPU sys | 84.3s |
| Peak spill | 0 B |
</details>
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] bench: isolate continuous partial flushing against main [datafusion]
Posted by "codecov-commenter (via GitHub)" <gi...@apache.org>.
codecov-commenter commented on PR #26098:
URL: https://github.com/apache/datafusion/pull/26098#issuecomment-6031676928
## [Codecov](https://app.codecov.io/gh/apache/datafusion/pull/26098?dropdown=coverage&src=pr&el=h1&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache) Report
:x: Patch coverage is `88.65385%` with `118 lines` in your changes missing coverage. Please review.
:white_check_mark: Project coverage is 82.73%. Comparing base ([`22a0fdb`](https://app.codecov.io/gh/apache/datafusion/commit/22a0fdb64614f4d751ed0009dbe1ac850ec79282?dropdown=coverage&el=desc&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache)) to head ([`fe0866a`](https://app.codecov.io/gh/apache/datafusion/commit/fe0866a91902f5a835be6a68713a78c0855c56b1?dropdown=coverage&el=desc&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache)).
:warning: Report is 46 commits behind head on main.
| [Files with missing lines](https://app.codecov.io/gh/apache/datafusion/pull/26098?dropdown=coverage&src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache) | Patch % | Lines |
|---|---|---|
| [...fusion/physical-plan/src/aggregates/hash\_stream.rs](https://app.codecov.io/gh/apache/datafusion/pull/26098?src=pr&el=tree&filepath=datafusion%2Fphysical-plan%2Fsrc%2Faggregates%2Fhash_stream.rs&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache#diff-ZGF0YWZ1c2lvbi9waHlzaWNhbC1wbGFuL3NyYy9hZ2dyZWdhdGVzL2hhc2hfc3RyZWFtLnJz) | 90.04% | [9 Missing and 35 partials :warning: ](https://app.codecov.io/gh/apache/datafusion/pull/26098?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache) |
| [...sion/physical-plan/src/aggregates/final\_buckets.rs](https://app.codecov.io/gh/apache/datafusion/pull/26098?src=pr&el=tree&filepath=datafusion%2Fphysical-plan%2Fsrc%2Faggregates%2Ffinal_buckets.rs&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache#diff-ZGF0YWZ1c2lvbi9waHlzaWNhbC1wbGFuL3NyYy9hZ2dyZWdhdGVzL2ZpbmFsX2J1Y2tldHMucnM=) | 86.69% | [8 Missing and 29 partials :warning: ](https://app.codecov.io/gh/apache/datafusion/pull/26098?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache) |
| [...ysical-plan/src/aggregates/bucketed\_aggregation.rs](https://app.codecov.io/gh/apache/datafusion/pull/26098?src=pr&el=tree&filepath=datafusion%2Fphysical-plan%2Fsrc%2Faggregates%2Fbucketed_aggregation.rs&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache#diff-ZGF0YWZ1c2lvbi9waHlzaWNhbC1wbGFuL3NyYy9hZ2dyZWdhdGVzL2J1Y2tldGVkX2FnZ3JlZ2F0aW9uLnJz) | 85.33% | [10 Missing and 23 partials :warning: ](https://app.codecov.io/gh/apache/datafusion/pull/26098?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache) |
| [...src/aggregates/aggregate\_hash\_table/final\_table.rs](https://app.codecov.io/gh/apache/datafusion/pull/26098?src=pr&el=tree&filepath=datafusion%2Fphysical-plan%2Fsrc%2Faggregates%2Faggregate_hash_table%2Ffinal_table.rs&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache#diff-ZGF0YWZ1c2lvbi9waHlzaWNhbC1wbGFuL3NyYy9hZ2dyZWdhdGVzL2FnZ3JlZ2F0ZV9oYXNoX3RhYmxlL2ZpbmFsX3RhYmxlLnJz) | 90.00% | [3 Missing :warning: ](https://app.codecov.io/gh/apache/datafusion/pull/26098?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache) |
| [...plan/src/aggregates/aggregate\_hash\_table/common.rs](https://app.codecov.io/gh/apache/datafusion/pull/26098?src=pr&el=tree&filepath=datafusion%2Fphysical-plan%2Fsrc%2Faggregates%2Faggregate_hash_table%2Fcommon.rs&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache#diff-ZGF0YWZ1c2lvbi9waHlzaWNhbC1wbGFuL3NyYy9hZ2dyZWdhdGVzL2FnZ3JlZ2F0ZV9oYXNoX3RhYmxlL2NvbW1vbi5ycw==) | 98.33% | [0 Missing and 1 partial :warning: ](https://app.codecov.io/gh/apache/datafusion/pull/26098?src=pr&el=tree&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache) |
<details><summary>Additional details and impacted files</summary>
```diff
@@ Coverage Diff @@
## main #26098 +/- ##
==========================================
+ Coverage 82.64% 82.73% +0.09%
==========================================
Files 1147 1149 +2
Lines 445440 450009 +4569
Branches 445440 450009 +4569
==========================================
+ Hits 368123 372333 +4210
+ Misses 54970 54948 -22
- Partials 22347 22728 +381
```
</details>
[:umbrella: View full report in Codecov by Harness](https://app.codecov.io/gh/apache/datafusion/pull/26098?dropdown=coverage&src=pr&el=continue&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache).
:loudspeaker: Have feedback on the report? [Share it here](https://about.codecov.io/codecov-pr-comment-feedback/?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=apache).
<details><summary> :rocket: New features to boost your workflow: </summary>
- :snowflake: [Test Analytics](https://docs.codecov.com/docs/test-analytics): Detect flaky tests, report on failures, and find test suite problems.
- :package: [JS Bundle Analysis](https://docs.codecov.com/docs/javascript-bundle-analysis): Save yourself from yourself by tracking and limiting bundle sizes in JS merges.
</details>
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] bench: isolate continuous partial flushing against main [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #26098:
URL: https://github.com/apache/datafusion/pull/26098#issuecomment-6031576723
π€ Benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/26098#issuecomment-6031351170)
**Instance:** `c4a-highcpu-48` (12 vCPU / 65 GiB) | `Linux bench-c6031351170-3108-qg2hl 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux`
<details><summary>CPU Details (lscpu)</summary>
```
Architecture: aarch64
CPU op-mode(s): 64-bit
Byte Order: Little Endian
CPU(s): 48
On-line CPU(s) list: 0-47
Vendor ID: ARM
Model name: Neoverse-V2
Model: 1
Thread(s) per core: 1
Core(s) per cluster: 48
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: 3 MiB (48 instances)
L1i cache: 3 MiB (48 instances)
L2 cache: 96 MiB (48 instances)
L3 cache: 80 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0-47
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 codex/partial-flush-controls (fe0866a91902f5a835be6a68713a78c0855c56b1) to 7f0ab426eb84a967a93c42ee0bd6e928e49adeb6 [diff](https://github.com/apache/datafusion/compare/7f0ab426eb84a967a93c42ee0bd6e928e49adeb6..fe0866a91902f5a835be6a68713a78c0855c56b1)
<details><summary>Run configuration</summary>
```yaml
run benchmark clickbench_partitioned
env:
DATAFUSION_EXECUTION_HASH_AGGREGATE_BUCKET_COMPACTION: "false"
DATAFUSION_EXECUTION_HASH_AGGREGATE_BUCKET_THRESHOLD: "262144"
DATAFUSION_EXECUTION_HASH_AGGREGATE_DETECT_REPEATED_GROUPS: "false"
DATAFUSION_EXECUTION_HASH_AGGREGATE_FINAL_BUCKETS: "false"
baseline:
ref: "7f0ab426eb84a967a93c42ee0bd6e928e49adeb6"
```
</details>
Results will be posted here when complete
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] bench: isolate continuous partial flushing against main [datafusion]
Posted by "Rachelint (via GitHub)" <gi...@apache.org>.
Rachelint commented on PR #26098:
URL: https://github.com/apache/datafusion/pull/26098#issuecomment-6031351170
run benchmark clickbench_partitioned
baseline:
ref: 7f0ab426eb84a967a93c42ee0bd6e928e49adeb6
env:
DATAFUSION_EXECUTION_HASH_AGGREGATE_BUCKET_THRESHOLD: "262144"
DATAFUSION_EXECUTION_HASH_AGGREGATE_BUCKET_COMPACTION: "false"
DATAFUSION_EXECUTION_HASH_AGGREGATE_DETECT_REPEATED_GROUPS: "false"
DATAFUSION_EXECUTION_HASH_AGGREGATE_FINAL_BUCKETS: "false"
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] bench: isolate continuous partial flushing against main [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #26098:
URL: https://github.com/apache/datafusion/pull/26098#issuecomment-6031761915
π€ Benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/26098#issuecomment-6031699653)
**Instance:** `c4a-highcpu-48` (12 vCPU / 65 GiB) | `Linux bench-c6031699653-3110-ttzc8 6.12.94+ #1 SMP Fri Aug 21 08:00:16 UTC 2026 aarch64 GNU/Linux`
<details><summary>CPU Details (lscpu)</summary>
```
Architecture: aarch64
CPU op-mode(s): 64-bit
Byte Order: Little Endian
CPU(s): 48
On-line CPU(s) list: 0-47
Vendor ID: ARM
Model name: Neoverse-V2
Model: 1
Thread(s) per core: 1
Core(s) per cluster: 48
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: 3 MiB (48 instances)
L1i cache: 3 MiB (48 instances)
L2 cache: 96 MiB (48 instances)
L3 cache: 80 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0-47
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 codex/partial-flush-controls (fe0866a91902f5a835be6a68713a78c0855c56b1) to 7f0ab426eb84a967a93c42ee0bd6e928e49adeb6 [diff](https://github.com/apache/datafusion/compare/7f0ab426eb84a967a93c42ee0bd6e928e49adeb6..fe0866a91902f5a835be6a68713a78c0855c56b1)
<details><summary>Run configuration</summary>
```yaml
run benchmark clickbench_partitioned
env:
DATAFUSION_EXECUTION_HASH_AGGREGATE_BUCKET_COMPACTION: "false"
DATAFUSION_EXECUTION_HASH_AGGREGATE_BUCKET_THRESHOLD: "262144"
DATAFUSION_EXECUTION_HASH_AGGREGATE_DETECT_REPEATED_GROUPS: "false"
DATAFUSION_EXECUTION_HASH_AGGREGATE_FINAL_BUCKETS: "true"
baseline:
ref: "7f0ab426eb84a967a93c42ee0bd6e928e49adeb6"
```
</details>
Results will be posted here when complete
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] bench: isolate continuous partial flushing against main [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #26098:
URL: https://github.com/apache/datafusion/pull/26098#issuecomment-6031867677
π€ Benchmark completed (GKE) | [trigger](https://github.com/apache/datafusion/pull/26098#issuecomment-6031699653)
**Instance:** `c4a-highcpu-48` (12 vCPU / 65 GiB)
Comparing codex/partial-flush-controls (fe0866a91902f5a835be6a68713a78c0855c56b1) to 7f0ab426eb84a967a93c42ee0bd6e928e49adeb6 [diff](https://github.com/apache/datafusion/compare/7f0ab426eb84a967a93c42ee0bd6e928e49adeb6..fe0866a91902f5a835be6a68713a78c0855c56b1)
<details><summary>Run configuration</summary>
```yaml
run benchmark clickbench_partitioned
env:
DATAFUSION_EXECUTION_HASH_AGGREGATE_BUCKET_COMPACTION: "false"
DATAFUSION_EXECUTION_HASH_AGGREGATE_BUCKET_THRESHOLD: "262144"
DATAFUSION_EXECUTION_HASH_AGGREGATE_DETECT_REPEATED_GROUPS: "false"
DATAFUSION_EXECUTION_HASH_AGGREGATE_FINAL_BUCKETS: "true"
baseline:
ref: "7f0ab426eb84a967a93c42ee0bd6e928e49adeb6"
```
</details>
<details><summary>CPU Details (lscpu)</summary>
```
Architecture: aarch64
CPU op-mode(s): 64-bit
Byte Order: Little Endian
CPU(s): 48
On-line CPU(s) list: 0-47
Vendor ID: ARM
Model name: Neoverse-V2
Model: 1
Thread(s) per core: 1
Core(s) per cluster: 48
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: 3 MiB (48 instances)
L1i cache: 3 MiB (48 instances)
L2 cache: 96 MiB (48 instances)
L3 cache: 80 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0-47
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>
<details><summary>Details</summary>
<p>
```
Comparing HEAD and codex_partial-flush-controls
--------------------
Benchmark clickbench_partitioned.json
--------------------
βββββββββββββ³βββββββββββββ³βββββββββββββββββββββββββββββββ³ββββββββββββββββ
β Query β HEAD β codex_partial-flush-controls β Change β
β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
β QQuery 0 β 1.21 ms β 1.25 ms β no change β
β QQuery 1 β 11.74 ms β 11.56 ms β no change β
β QQuery 2 β 36.56 ms β 36.51 ms β no change β
β QQuery 3 β 30.86 ms β 31.10 ms β no change β
β QQuery 4 β 231.75 ms β 163.65 ms β +1.42x faster β
β QQuery 5 β 267.84 ms β 258.91 ms β no change β
β QQuery 6 β 1.30 ms β 1.32 ms β no change β
β QQuery 7 β 12.86 ms β 12.89 ms β no change β
β QQuery 8 β 324.79 ms β 264.69 ms β +1.23x faster β
β QQuery 9 β 467.63 ms β 475.57 ms β no change β
β QQuery 10 β 63.28 ms β 66.27 ms β no change β
β QQuery 11 β 73.84 ms β 77.14 ms β no change β
β QQuery 12 β 258.17 ms β 246.05 ms β no change β
β QQuery 13 β 356.94 ms β 400.35 ms β 1.12x slower β
β QQuery 14 β 274.79 ms β 264.41 ms β no change β
β QQuery 15 β 287.33 ms β 183.90 ms β +1.56x faster β
β QQuery 16 β 609.14 ms β 530.09 ms β +1.15x faster β
β QQuery 17 β 615.56 ms β 445.39 ms β +1.38x faster β
β QQuery 18 β 1260.61 ms β 981.50 ms β +1.28x faster β
β QQuery 19 β 27.26 ms β 27.46 ms β no change β
β QQuery 20 β 516.80 ms β 526.60 ms β no change β
β QQuery 21 β 516.10 ms β 514.39 ms β no change β
β QQuery 22 β 983.55 ms β 1019.63 ms β no change β
β QQuery 23 β 3058.41 ms β 3195.60 ms β no change β
β QQuery 24 β 40.36 ms β 39.91 ms β no change β
β QQuery 25 β 105.15 ms β 104.36 ms β no change β
β QQuery 26 β 40.68 ms β 40.33 ms β no change β
β QQuery 27 β 513.41 ms β 510.48 ms β no change β
β QQuery 28 β 2876.49 ms β 2880.14 ms β no change β
β QQuery 29 β 38.67 ms β 39.18 ms β no change β
β QQuery 30 β 309.02 ms β 259.97 ms β +1.19x faster β
β QQuery 31 β 273.11 ms β 250.81 ms β +1.09x faster β
β QQuery 32 β 1068.34 ms β 642.75 ms β +1.66x faster β
β QQuery 33 β 1572.63 ms β 1373.65 ms β +1.14x faster β
β QQuery 34 β 1593.09 ms β 1399.02 ms β +1.14x faster β
β QQuery 35 β 299.81 ms β 206.20 ms β +1.45x faster β
β QQuery 36 β 66.61 ms β 67.44 ms β no change β
β QQuery 37 β 34.81 ms β 36.06 ms β no change β
β QQuery 38 β 43.05 ms β 42.21 ms β no change β
β QQuery 39 β 145.95 ms β 142.61 ms β no change β
β QQuery 40 β 13.96 ms β 14.29 ms β no change β
β QQuery 41 β 13.46 ms β 13.81 ms β no change β
β QQuery 42 β 12.91 ms β 12.99 ms β no change β
βββββββββββββ΄βββββββββββββ΄βββββββββββββββββββββββββββββββ΄ββββββββββββββββ
βββββββββββββββββββββββββββββββββββββββββββββββ³βββββββββββββ
β Benchmark Summary β β
β‘βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
β Total Time (HEAD) β 19349.85ms β
β Total Time (codex_partial-flush-controls) β 17812.43ms β
β Average Time (HEAD) β 450.00ms β
β Average Time (codex_partial-flush-controls) β 414.24ms β
β Queries Faster β 12 β
β Queries Slower β 1 β
β Queries with No Change β 30 β
β Queries with Failure β 0 β
βββββββββββββββββββββββββββββββββββββββββββββββ΄βββββββββββββ
Distribution per query (min / mean Β±stddev / max):
Comparing HEAD and codex_partial-flush-controls
--------------------
Benchmark clickbench_partitioned.json
--------------------
βββββββββββββ³βββββββββββββββββββββββββββββββββββββββββ³βββββββββββββββββββββββββββββββββββββββββ³ββββββββββββββββ
β Query β HEAD β codex_partial-flush-controls β Change β
β‘ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
β QQuery 0 β 1.21 / 4.00 Β±5.49 / 14.98 ms β 1.25 / 4.04 Β±5.51 / 15.07 ms β no change β
β QQuery 1 β 11.74 / 11.85 Β±0.07 / 11.92 ms β 11.56 / 11.75 Β±0.12 / 11.92 ms β no change β
β QQuery 2 β 36.56 / 36.66 Β±0.08 / 36.80 ms β 36.51 / 36.69 Β±0.17 / 36.90 ms β no change β
β QQuery 3 β 30.86 / 31.34 Β±0.57 / 32.44 ms β 31.10 / 31.26 Β±0.12 / 31.42 ms β no change β
β QQuery 4 β 231.75 / 236.03 Β±2.57 / 239.09 ms β 163.65 / 165.34 Β±1.23 / 167.43 ms β +1.43x faster β
β QQuery 5 β 267.84 / 274.20 Β±4.00 / 279.54 ms β 258.91 / 265.83 Β±3.57 / 268.93 ms β no change β
β QQuery 6 β 1.30 / 1.44 Β±0.22 / 1.87 ms β 1.32 / 1.46 Β±0.22 / 1.89 ms β no change β
β QQuery 7 β 12.86 / 13.01 Β±0.13 / 13.20 ms β 12.89 / 12.95 Β±0.04 / 12.99 ms β no change β
β QQuery 8 β 324.79 / 331.55 Β±5.04 / 338.83 ms β 264.69 / 268.39 Β±3.01 / 273.64 ms β +1.24x faster β
β QQuery 9 β 467.63 / 476.73 Β±4.98 / 482.56 ms β 475.57 / 481.84 Β±6.62 / 494.57 ms β no change β
β QQuery 10 β 63.28 / 63.78 Β±0.71 / 65.18 ms β 66.27 / 67.76 Β±1.34 / 70.22 ms β 1.06x slower β
β QQuery 11 β 73.84 / 74.92 Β±0.68 / 75.72 ms β 77.14 / 77.64 Β±0.56 / 78.67 ms β no change β
β QQuery 12 β 258.17 / 264.68 Β±3.94 / 270.49 ms β 246.05 / 266.27 Β±24.87 / 314.62 ms β no change β
β QQuery 13 β 356.94 / 369.99 Β±14.34 / 392.15 ms β 400.35 / 406.23 Β±9.85 / 425.90 ms β 1.10x slower β
β QQuery 14 β 274.79 / 286.69 Β±13.48 / 312.06 ms β 264.41 / 271.54 Β±5.87 / 282.06 ms β +1.06x faster β
β QQuery 15 β 287.33 / 293.95 Β±4.17 / 299.54 ms β 183.90 / 188.03 Β±3.45 / 193.92 ms β +1.56x faster β
β QQuery 16 β 609.14 / 616.89 Β±4.47 / 622.75 ms β 530.09 / 542.27 Β±13.50 / 567.60 ms β +1.14x faster β
β QQuery 17 β 615.56 / 624.71 Β±8.42 / 638.61 ms β 445.39 / 463.22 Β±26.84 / 516.19 ms β +1.35x faster β
β QQuery 18 β 1260.61 / 1291.83 Β±22.97 / 1321.79 ms β 981.50 / 991.24 Β±7.03 / 999.53 ms β +1.30x faster β
β QQuery 19 β 27.26 / 32.00 Β±8.65 / 49.26 ms β 27.46 / 27.78 Β±0.17 / 27.96 ms β +1.15x faster β
β QQuery 20 β 516.80 / 529.29 Β±12.89 / 552.18 ms β 526.60 / 558.51 Β±54.76 / 667.80 ms β 1.06x slower β
β QQuery 21 β 516.10 / 520.44 Β±6.30 / 532.85 ms β 514.39 / 536.83 Β±12.10 / 548.67 ms β no change β
β QQuery 22 β 983.55 / 991.09 Β±6.76 / 1002.48 ms β 1019.63 / 1037.15 Β±10.48 / 1048.95 ms β no change β
β QQuery 23 β 3058.41 / 3272.16 Β±138.38 / 3467.25 ms β 3195.60 / 3398.83 Β±177.35 / 3629.46 ms β no change β
β QQuery 24 β 40.36 / 77.71 Β±36.55 / 123.72 ms β 39.91 / 130.52 Β±154.06 / 435.71 ms β 1.68x slower β
β QQuery 25 β 105.15 / 146.93 Β±73.45 / 292.97 ms β 104.36 / 154.67 Β±95.27 / 345.14 ms β 1.05x slower β
β QQuery 26 β 40.68 / 41.64 Β±0.85 / 43.18 ms β 40.33 / 42.63 Β±3.84 / 50.27 ms β no change β
β QQuery 27 β 513.41 / 522.42 Β±5.40 / 528.83 ms β 510.48 / 517.94 Β±8.29 / 532.71 ms β no change β
β QQuery 28 β 2876.49 / 2900.81 Β±36.07 / 2972.37 ms β 2880.14 / 2921.30 Β±50.72 / 3020.85 ms β no change β
β QQuery 29 β 38.67 / 43.03 Β±7.72 / 58.44 ms β 39.18 / 41.62 Β±3.20 / 47.85 ms β no change β
β QQuery 30 β 309.02 / 312.08 Β±1.83 / 314.10 ms β 259.97 / 265.94 Β±4.45 / 272.79 ms β +1.17x faster β
β QQuery 31 β 273.11 / 291.63 Β±10.76 / 304.07 ms β 250.81 / 259.59 Β±5.22 / 265.48 ms β +1.12x faster β
β QQuery 32 β 1068.34 / 1165.79 Β±75.37 / 1245.28 ms β 642.75 / 694.67 Β±46.57 / 780.25 ms β +1.68x faster β
β QQuery 33 β 1572.63 / 1613.64 Β±33.78 / 1656.94 ms β 1373.65 / 1444.81 Β±38.38 / 1482.27 ms β +1.12x faster β
β QQuery 34 β 1593.09 / 1618.36 Β±20.31 / 1654.79 ms β 1399.02 / 1476.48 Β±53.04 / 1560.60 ms β +1.10x faster β
β QQuery 35 β 299.81 / 338.55 Β±62.35 / 462.88 ms β 206.20 / 260.61 Β±86.39 / 431.11 ms β +1.30x faster β
β QQuery 36 β 66.61 / 74.10 Β±7.81 / 89.02 ms β 67.44 / 73.30 Β±5.85 / 83.22 ms β no change β
β QQuery 37 β 34.81 / 39.59 Β±5.09 / 47.29 ms β 36.06 / 38.20 Β±2.42 / 41.18 ms β no change β
β QQuery 38 β 43.05 / 48.30 Β±5.36 / 56.66 ms β 42.21 / 45.01 Β±2.39 / 48.66 ms β +1.07x faster β
β QQuery 39 β 145.95 / 153.75 Β±6.25 / 160.98 ms β 142.61 / 146.89 Β±3.04 / 150.74 ms β no change β
β QQuery 40 β 13.96 / 14.15 Β±0.15 / 14.34 ms β 14.29 / 14.58 Β±0.43 / 15.42 ms β no change β
β QQuery 41 β 13.46 / 18.43 Β±6.10 / 27.15 ms β 13.81 / 16.57 Β±3.26 / 21.41 ms β +1.11x faster β
β QQuery 42 β 12.91 / 13.12 Β±0.16 / 13.31 ms β 12.99 / 13.26 Β±0.27 / 13.71 ms β no change β
βββββββββββββ΄βββββββββββββββββββββββββββββββββββββββββ΄βββββββββββββββββββββββββββββββββββββββββ΄ββββββββββββββββ
βββββββββββββββββββββββββββββββββββββββββββββββ³βββββββββββββ
β Benchmark Summary β β
β‘βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ©
β Total Time (HEAD) β 20083.25ms β
β Total Time (codex_partial-flush-controls) β 18671.40ms β
β Average Time (HEAD) β 467.05ms β
β Average Time (codex_partial-flush-controls) β 434.22ms β
β Queries Faster β 16 β
β Queries Slower β 5 β
β Queries with No Change β 22 β
β Queries with Failure β 0 β
βββββββββββββββββββββββββββββββββββββββββββββββ΄βββββββββββββ
```
</p>
</details>
<details><summary>Resource Usage</summary>
**clickbench_partitioned β base (merge-base)**
| Metric | Value |
|--------|-------|
| Wall time | 105.0s |
| Peak memory | 18.1 GiB |
| Avg memory | 5.8 GiB |
| CPU user | 991.7s |
| CPU sys | 97.5s |
| Peak spill | 0 B |
**clickbench_partitioned β branch**
| Metric | Value |
|--------|-------|
| Wall time | 95.0s |
| Peak memory | 17.5 GiB |
| Avg memory | 5.3 GiB |
| CPU user | 893.1s |
| CPU sys | 88.3s |
| Peak spill | 0 B |
</details>
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org