zuston opened a new pull request, #2578: URL: https://github.com/apache/uniffle/pull/2578
### What changes were proposed in this pull request? This PR is to use the space sensitive protobuf data structure to store the partitions -> servers mapping, that is to reduce the rpc cost ### Why are the changes needed? This is the part of PR for the #2568. In large-scale Spark jobs, the number of partitions can reach up to 20K, whereas the number of assigned shuffle servers remains smaller than the total number of nodes in the Uniffle cluster. Prior to this PR, both the driver and the client (when reassignment was enabled) required substantial memory for RPC transfers, which could significantly increase the frequency of driver garbage collection. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? Unit tests. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@uniffle.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@uniffle.apache.org For additional commands, e-mail: issues-h...@uniffle.apache.org
