Thanks Sean for your reply Do you have any plan to make it configurable of TSM compaction to prevent from disk full early and write/query blocking during full compaction?
2016년 9월 23일 금요일 오전 11시 33분 46초 UTC+9, Jason Park 님의 말: > > Hi there, > > I'm testing for feasibility of influxdb performance. > > DB schema as following. > Measurements : 100 > Series : 30M with one Tag and five fields. > > Traffic generation : 30K write points data per second with wrk2 HTTP load > generation tool. > Used influxdb default configuration but max cache memory was 50G. > 32 Cores CPU, 512G physical memory, 2T SSD disk > > As you can see the attachment, InfluxDB shard size shows big saw teeth, > looks like fully compaction every 7~8 hour. > I hope that I can make it smaller it. *Is it configurable?* > > Actually it seems affect write and query blocking because InfluxDB perform > any compaction after that Go routines is increased up to 20 times, > at that time the 'PointReq' is 0 for long time up to 15 Min. > > Thank you in advance > -- Remember to include the InfluxDB version number with all issue reports --- You received this message because you are subscribed to the Google Groups "InfluxDB" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/influxdb. To view this discussion on the web visit https://groups.google.com/d/msgid/influxdb/959cdfff-d4ca-428b-941d-de44b1ac0a79%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
