Support variable query repetition counts - #1358
Open
caetanosauer wants to merge 1 commit into
Open
Conversation
For the fastest systems in this benchmark, all queries run in under a second for the hot runs. This leads to a high variability between different executions on the exact same system and hardware. To alleviate this problem, this commit enhances the dashboard and associated metric scores to support an arbitrary number of repetitions instead of the fixed 3. It aligns with the fact that the execution scripts already support this through `BENCH_TRIES` (see `lib/benchmark-common.sh`). Disclaimer: This leads to better results than currently published for our system (Hyper), which we includ in this PR for the `c7a.metal-48xl` machine. We encourage other systems to do the same so we can have more stable results overall.
Contributor
Author
|
@alexey-milovidov Curious to hear your thoughts on this one |
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
For the fastest systems in this benchmark, all queries run in under a second for the hot runs.
This leads to a high variability between different executions on the exact same system and hardware.
To alleviate this problem, this commit enhances the dashboard and associated metric scores to support
an arbitrary number of repetitions instead of the fixed 3. It aligns with the fact that the execution
scripts already support this through
BENCH_TRIES(seelib/benchmark-common.sh).Disclaimer: This leads to better results than currently published for our system (Hyper), which we
includ in this PR for the
c7a.metal-48xlmachine. We encourageother systems to do the same so we can have more stable results overall.
This commit does not change the aggregate function, i.e., it still takes the min of the repetitions.
We might consider using median for more repetitions, but I left that out of scope here.
Note: this also addresses issue #934