Repository navigation
rules(performance): parallel splits must scale to the largest host's thread count - #28
Merged
Merged
Conversation
…thread count Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds a Parallel Scaling section to
rules/common/performance.md: a parallel split must keep working when the thread count grows to the largest host (64 cores and more), not only at the 4 to 8 threads of a development machine.It lists the failure modes to check before measuring a new or changed parallel region: a shared operand read once per task (traffic
p × size),O(p)partial results with anO(p)combine, fixed task caps or split counts, thresholds that ignorep, contention that grows withp, and multiple cache domains. It asks for a comparison with the reference implementation (reference LAPACK / OpenBLAS for dense linear algebra), scaling evidence at 1, 8 and the largest thread count, traffic numbers when the claim is about traffic, and judging by mechanism rather than tuning constants to one host.Motivation: tensor4all/faer#90, where the block Householder apply's column-only split re-reads the whole basis once per task, so its traffic grows with the thread count.
🤖 Generated with Claude Code