nickva commented on PR #5625: URL: https://github.com/apache/couchdb/pull/5625#issuecomment-3209109828
> Nice work! Is there a break even point where things get worser? Can you run those tests on smaller clusters too? I'll try to find a smaller cluster but sometime setting it up my not work as well as our dedicated perf cluster. I could run some more benchmarks on my ubuntu laptop, that could be an example of a small machine. > Is there a break even point where things get worser? I'd imagine if we shuffled every node through the cache, things would get worse. As we'd fill it up with nodes used only a times. We see the distribution of counts in one of the above comments, at level 2 or 3 we may have only 40-50 kp nodes and one level lower there are thousands. The `full` metric can sort of help determine if the cache size may need to be bumped or if the depth may need to be reduced to make it more effective. -- 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: notifications-unsubscr...@couchdb.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
