graphite-dev team mailing list archive

[Yousef Fattal <question663652@xxxxxxxxxxxxxxxxxxxxx>]

New question #663652 on Graphite:
https://answers.launchpad.net/graphite/+question/663652

i am using graphite as a data source in grafana, I have no issues with the server performance on all the different dashboards and queries but we are now testing a new map panel that can have the metrics plotted on it.


the panel have 5K+  metrics plotted and when it refreshes for the new metrics from graphite, the httpd process on graphite showing that it's demanding 130%  %CPU via the top command ( running graphite on redhat 7.4 by the way)

what's interesting is that only one of the cores is being utilized at %60 percent at my server when the refresh happens.

so my question is there anyway to make that particulate query get spanned on multiple process/threads so it can utilize more of the 16 cores on my server ?  that may also enhance the time the query take to refresh on grafana which is ~15 seconds.

any other suggestion is highly appreciated to go around this

thank you for the help

-- 
You received this question notification because your team graphite-dev
is an answer contact for Graphite.