Shekharrajak opened a new pull request, #22835:
URL: https://github.com/apache/kafka/pull/22835

   PersisterStateManager coalesces many partitions into one 
WRITE/READ/…_SHARE_GROUP_STATE RPC. On completion, every per-partition handler 
re-scans the full combined response to find its own slice — **N handlers  *  
O(N) scan = O(N ^ 2),** run on the single send thread, stalling the send loop.
   
   This PR adds the shared plumbing to demultiplex the combined response once 
into a topicId  ->  partition -> result index (indexByTopicPartition + base 
buildResultIndex/lookupPartitionResult, built in the send 
   loop and shared across handlers), and converts WriteStateHandler to an O(1) 
lookup.
   
   
   Foundation PR  for tthe ReadStateHandler / ReadStateSummaryHandler / 
InitializeStateHandler / DeleteStateHandler PRs will be built on this.


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to