Sharded probe table #3556
Draft
Sharded probe table #3556
Conversation
This file contains 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
colin-ho
added a commit
that referenced
this pull request
Dec 17, 2024
Optimize swordfish grouped aggs for high cardinality groups ### Approach There's 3 strategies for grouped aggs: 1. Partition each input morsel into `N` partitions, then do a partial agg. (good for high cardinality). 2. Do a partial agg, then partition into `N` partitions. (good for low cardinality). Can be optimized with #3556 3. Partition only, no partial agg. (only for map_groups, which has no partial agg). ### Notes on alternative approaches - Distributing partitions across workers (i.e. having each worker being responsible for accumulating only one partition) is much slower for low cardinality aggs (TPCH Q1 would have been 1.5x slower). This is because most of the work will end up being on only a few workers, reducing parallelism. - Simply partitioning the input and then only aggregating at the end works well with higher cardinality, but low cardinality takes a hit. (TPCH Q1 would have been 2.5x slower). - Probe Table approach was much slower, due to many calls to the multi-table dyn comparator. It was also much more complex to implement. ### Benchmarks [MrPowers Benchmarks](https://github.com/MrPowers/mrpowers-benchmarks) results (seconds, lower is better). | Query | this PR | Pyrunner | Current swordfish | |-------|---------|----------|-------------------| | q1 | 0.285720| 0.768858 | 0.356499 | | q2 | 4.780064| 6.122199 | 53.340565 | | q3 | 2.201079| 3.922857 | 16.935125 | | q4 | 0.313106| 0.545192 | 0.335541 | | q5 | 1.618228| 2.889354 | 10.665339 | | q7 | 2.087872| 3.856998 | 16.072660 | | q10 | 6.306756| 8.173738 | 53.800501 | --------- Co-authored-by: EC2 Default User <[email protected]> Co-authored-by: Colin Ho <[email protected]> Co-authored-by: EC2 Default User <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
No description provided.