You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have searched in the issues and found no similar issues.
What would you like to be improved?
Assuming there are 500 running APPs, each APP has an average of 10 shuffleIds, and each shuffleId has 1000 partitions, then the size of partitionsOfStorage in LocalStorageManager is about 5000000. The cleanupStorageSelectionCache method can take 500ms on average (sometimes it took a longer time when the size of partitionsOfStorage is larger). So it should be optimized.
Also assuming we complete 10,000 APPs in a day, if we use the AppPurgeEvent event to release resources, it would take approximately 10,000 * 500ms = 5000s. But if we use the ShufflePurgeEvent event to release resources, then it would take 10,000 * 500ms * 10 = 50,000s, which is quite crazy.
How should we improve?
Optimize cleanupStorageSelectionCache method in LocalStorageManager.
Fix bug when release resource using ShufflePurgeEvent event.
Are you willing to submit PR?
Yes I am willing to submit a PR!
The text was updated successfully, but these errors were encountered:
…thod in LocalStorageManager. (#1210)
### What changes were proposed in this pull request?
Optimize cleanupStorageSelectionCache method in LocalStorageManager.
### Why are the changes needed?
Fix: #1209
### Does this PR introduce _any_ user-facing change?
No.
### How was this patch tested?
Existing UTs.
Code of Conduct
Search before asking
What would you like to be improved?
Assuming there are 500 running APPs, each APP has an average of 10 shuffleIds, and each shuffleId has 1000 partitions, then the size of partitionsOfStorage in LocalStorageManager is about 5000000. The cleanupStorageSelectionCache method can take 500ms on average (sometimes it took a longer time when the size of partitionsOfStorage is larger). So it should be optimized.
Also assuming we complete 10,000 APPs in a day, if we use the AppPurgeEvent event to release resources, it would take approximately 10,000 * 500ms = 5000s. But if we use the ShufflePurgeEvent event to release resources, then it would take 10,000 * 500ms * 10 = 50,000s, which is quite crazy.
How should we improve?
Are you willing to submit PR?
The text was updated successfully, but these errors were encountered: