-
Notifications
You must be signed in to change notification settings - Fork 1.1k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Memory arbitration related code refactor #11168
Conversation
This pull request was exported from Phabricator. Differential Revision: D63815963 |
✅ Deploy Preview for meta-velox canceled.
|
Summary: Update participant operator methods to better support global arbitration. Remove stats from existing arbitrator stats which are not used much in production monitoring to simplify code Add memory arbitration context type to differentiate local arbitration context and global arbitration context. The latter doesn't have an associated request memory poll as in global arbitration optimization Rename ScopedMemoryPoolArbitrationContext to MemoryPoolArbitrationSection as we have too many context things in memory code space Add metrics used by global arbitration optimization and remove the ones not used much in production monitoring Differential Revision: D63815963
2e7f6da
to
c7630fd
Compare
This pull request was exported from Phabricator. Differential Revision: D63815963 |
Summary: Update participant operator methods to better support global arbitration. Remove stats from existing arbitrator stats which are not used much in production monitoring to simplify code Add memory arbitration context type to differentiate local arbitration context and global arbitration context. The latter doesn't have an associated request memory poll as in global arbitration optimization Rename ScopedMemoryPoolArbitrationContext to MemoryPoolArbitrationSection as we have too many context things in memory code space Add metrics used by global arbitration optimization and remove the ones not used much in production monitoring Reviewed By: tanjialiang Differential Revision: D63815963
c7630fd
to
136e3a6
Compare
This pull request was exported from Phabricator. Differential Revision: D63815963 |
This pull request has been merged in 7b6ef14. |
Conbench analyzed the 1 benchmark run on commit There were no benchmark performance regressions. 🎉 The full Conbench report has more details. |
// The time that waits for global arbitration queue. | ||
uint64_t globalArbitrationWaitTimeUs_{0}; | ||
// The time that starts global arbitration wait | ||
uint64_t globalArbitrationStartTimeMs_{}; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
init to 0
slowCapacityGrowRatio, | ||
minFreeCapacity, | ||
succinctBytes(minFreeCapacity), |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can we add memoryPool back to prefix? It will help with readability. Also could you please make sure to make followup changes to all the same meaning name across the codebase to make it consistent?
Summary:
Update participant operator methods to better support global arbitration.
Remove stats from existing arbitrator stats which are not used much in production monitoring to simplify code
Add memory arbitration context type to differentiate local arbitration context and global arbitration context. The latter
doesn't have an associated request memory poll as in global arbitration optimization
Rename ScopedMemoryPoolArbitrationContext to MemoryPoolArbitrationSection as we have too many context things
in memory code space
Add metrics used by global arbitration optimization and remove the ones not used much in production monitoring
Differential Revision: D63815963