Alexey Serbin has posted comments on this change. ( http://gerrit.cloudera.org:8080/19197 )
Change subject: KUDU-3406: add --memory_flush_prioritization_prob flag ...................................................................... Patch Set 1: (1 comment) http://gerrit.cloudera.org:8080/#/c/19197/1/src/kudu/util/maintenance_manager.cc File src/kudu/util/maintenance_manager.cc: http://gerrit.cloudera.org:8080/#/c/19197/1/src/kudu/util/maintenance_manager.cc@538 PS1, Line 538: if (!best_perf_improvement_op || best_perf_improvement <= 0 || > Out of curiosity, is it possible that best_perf_improvement_op itself is a You are right -- that's what I was also looking at. I also saw an evidence of this in the wild (logs of a tablet server in a busy cluster): there are FlushDeltaMemStoresOp operations with very high perf score (e.g., 530). That's an important point, even if this can be somewhat mitigated by the higher value for --flush_threshold_mb. Yes, I think it's better to update this heuristic to allow for more compactions to take place. -- To view, visit http://gerrit.cloudera.org:8080/19197 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I07e33bcebe7284da20e57dfbc9d87070f92fd6f3 Gerrit-Change-Number: 19197 Gerrit-PatchSet: 1 Gerrit-Owner: Alexey Serbin <[email protected]> Gerrit-Reviewer: Abhishek Chennaka <[email protected]> Gerrit-Reviewer: Alexey Serbin <[email protected]> Gerrit-Reviewer: Ashwani Raina <[email protected]> Gerrit-Reviewer: Kudu Jenkins (120) Gerrit-Comment-Date: Fri, 04 Nov 2022 05:27:58 +0000 Gerrit-HasComments: Yes
