kocolosk commented on issue #1065: [Question] Periodic performance degradation URL: https://github.com/apache/couchdb/issues/1065#issuecomment-351756427 OK, it may be the case that the server is triggering compactions and the competition for resources is what is slowing you down in those intervals. When the compaction finishes your throughput is restored. You can determine if this is the case by checking the Active Tasks in the dashboard (or use the `_active_tasks` endpoint). Changing the `check_interval` might defer things a bit, but you'd still have the possibility of a compaction running during your benchmark. There are many other settings you can apply to tune the compactor's resource utilization: http://docs.couchdb.org/en/2.1.1/config/compaction.html You could also disable compaction during the test, but note that compaction does need to run periodically even if your database never updates individual documents. Note that compaction is a separate process from the one generating the `pending_changes` messages. I think those are less likely to be related to the changes in throughput personally. I'm going to go ahead and close this issue for now but by all means re-open if you have further questions :)
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
