[CELEBORN-1497] Optimize read buffer dispatcher to retain buffers until worker has memory pressure
What changes were proposed in this pull request?
To improve read buffer dispatcher performance.
Why are the changes needed?
In map partition scenario, there can be lots of buffer requests which performance issue in allocateing memory.
Does this PR introduce any user-facing change?
NO.
How was this patch tested?
GA and Cluster Test.
This PR is stale because it has been open 20 days with no activity. Remove stale label or comment or this will be closed in 10 days.
This issue was closed because it has been staled for 10 days with no activity.
This PR is stale because it has been open 20 days with no activity. Remove stale label or comment or this will be closed in 10 days.
This issue was closed because it has been staled for 10 days with no activity.