Scale cross-commit benchmark stats analysis to full sweeps - #6059
Open
q10 wants to merge 1 commit into
Open
Conversation
Summary: Make the benchmark analysis pipeline handle full-matrix cross-commit sweeps that produce tens of thousands of per-kernel trace rows. - aggregate_config_stats.py: add opt-in --group-by-pattern. Emits one row per real (config, dispatched --kernel-pattern group) measurement plus the per-config (total), instead of the config x distinct-kernel_name cross-product that was ~98% count=0 padding. Default behavior unchanged. - compute_stats.py: replace the O(keys x rows) nested key-match scan with an O(rows) first-occurrence key index (byte-identical output). Cuts full-sweep runtime from hours (timeout) to seconds. - run_benchmark.sh / extract_kernel_stats.py: pass --group-by-pattern so the canonical and OSS pipelines emit compact per-measurement CSVs. Reviewed By: henrylhtsang Differential Revision: D113330675
Contributor
|
@q10 has exported this pull request. If you are a Meta employee, you can view the originating Diff in D113330675. |
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary:
Make the benchmark analysis pipeline handle full-matrix cross-commit sweeps
that produce tens of thousands of per-kernel trace rows.
real (config, dispatched --kernel-pattern group) measurement plus the
per-config (total), instead of the config x distinct-kernel_name
cross-product that was ~98% count=0 padding. Default behavior unchanged.
O(rows) first-occurrence key index (byte-identical output). Cuts full-sweep
runtime from hours (timeout) to seconds.
canonical and OSS pipelines emit compact per-measurement CSVs.
Reviewed By: henrylhtsang
Differential Revision: D113330675