[PoC] perf: optimize group-only group-by case for primitive cases (clickbench q4)#22145
Draft
waynexia wants to merge 6 commits into
Draft
[PoC] perf: optimize group-only group-by case for primitive cases (clickbench q4)#22145waynexia wants to merge 6 commits into
waynexia wants to merge 6 commits into
Conversation
Signed-off-by: Ruihang Xia <waynestxia@gmail.com>
github-actions
Bot
added
logical-expr
Contributor
|
run benchmarks |
|
Benchmark for this request failed. Last 20 lines of output: Click to expandFile an issue against this benchmark runner |
2 similar comments
|
Benchmark for this request failed. Last 20 lines of output: Click to expandFile 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 |
Signed-off-by: Ruihang Xia <waynestxia@gmail.com>
Contributor
|
run benchmarks |
|
Benchmark for this request failed. Last 20 lines of output: Click to expandFile an issue against this benchmark runner |
2 similar comments
|
Benchmark for this request failed. Last 20 lines of output: Click to expandFile 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 |
Member
Author
|
run benchmarks |
|
Hi @waynexia, thanks for the request (#22145 (comment)). Only whitelisted users can trigger benchmarks. Allowed users: Dandandan, Fokko, Jefffrey, Omega359, 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 |
Contributor
|
run benchmarks |
|
🤖 Benchmark running (GKE) | trigger CPU Details (lscpu)Comparing q4 (65980f0) to 7f2f78d (merge-base) diff using: tpch File an issue against this benchmark runner |
|
🤖 Benchmark running (GKE) | trigger CPU Details (lscpu)Comparing q4 (65980f0) to 7f2f78d (merge-base) diff using: clickbench_partitioned File an issue against this benchmark runner |
|
🤖 Benchmark running (GKE) | trigger CPU Details (lscpu)Comparing q4 (65980f0) to 7f2f78d (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 Usagetpcds — base (merge-base)
tpcds — branch
File an issue against this benchmark runner |
|
🤖 Benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usageclickbench_partitioned — base (merge-base)
clickbench_partitioned — branch
File an issue against this benchmark runner |
Member
Author
|
Hmmm, the result does not seem good 🤔 Let me dig deeper |
Contributor
Seems like the query is faster though (and memory is ~halved) so seems like a great optimization - I expect it for higher cardinality |
Contributor
|
run benchmark clickbench_partitioned clickbench_extended |
|
🤖 Benchmark running (GKE) | trigger CPU Details (lscpu)Comparing q4 (65980f0) to 7f2f78d (merge-base) diff using: clickbench_partitioned File an issue against this benchmark runner |
|
🤖 Benchmark running (GKE) | trigger CPU Details (lscpu)Comparing q4 (65980f0) to 7f2f78d (merge-base) diff using: clickbench_extended File an issue against this benchmark runner |
Dandandan
reviewed
May 14, 2026
| /// is obvious in high cardinality group by situation. | ||
| /// More details can see: | ||
| /// <https://github.com/apache/datafusion/issues/15961> | ||
| map: HashTable<(usize, u64)>, |
Contributor
There was a problem hiding this comment.
I found HashTable<(usize, T::Native)> also to be a bit faster than this current approach (and also saving memory). We can build the Vec<T::Native> when emitting data.
Could be part of some future PR.
|
🤖 Benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usageclickbench_partitioned — base (merge-base)
clickbench_partitioned — branch
File an issue against this benchmark runner |
|
🤖 Benchmark completed (GKE) | trigger Instance: CPU Details (lscpu)Details
Resource Usageclickbench_extended — base (merge-base)
clickbench_extended — branch
File an issue against this benchmark runner |
Member
Author
Some queries regress a lot... |
Contributor
This is likely noise, best is to look at the lowest reported times in those cases. |
Member
Author
|
I rerun it use an aws c6a.4xlarge instance. Improvement on q4 is reproduced, but still has lots of regressions and some of them seem to be real. I'll verify them on my local environment next Details |
Member
Author
|
rerun another round and those regression disappear 😼 |
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
3 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
This is an (agent made) PoC branch for clickbench q4
SELECT COUNT(DISTINCT UserID) FROM hits;. In my test environment it's about 25% faster thanmain.The basic idea is to skip maintaining group-id when there is no requirement of it, e.g., for group-by cases without any accumulator.
Other findings on this experiment:
SELECT COUNT(DISTINCT SearchPhrase) FROM hits;where the key is string type, the group calculation is not the critical pathWhat changes are included in this PR?
Are these changes tested?
Are there any user-facing changes?