keith-turner commented on issue #4978: URL: https://github.com/apache/accumulo/issues/4978#issuecomment-2414298407
The goal of this proposed change is to reduce the overall latency of compactions. It would be good to have metrics that help understand what this latency is in a running system like #4980. Would be nice to be able to compute a ratio like (compaction_run_time)/(compaction_admin_overhead+compaction_run_time). The compaction_admin_overhead is reserving and commiting compactions and any other time not related to running the actual compaction. The closer this ration is to 1 the more efficient the system is, the further it is from one the less efficient and that could indicate a bug or transient runtime issue that needs investigation. I am going to look into adding metrics to compute this ratio or see if existing metrics could be used. The metrics work should probably precede this work and be used to inform any potential work on this issue. -- 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]
