Disk yes, memory no. Prometheus uses mmap to read the on-disk data. On Fri, Jun 19, 2020 at 12:31 AM [email protected] <[email protected]> wrote:
> But would is there memory and disk usage overhead? > четверг, 18 июня 2020 г. в 23:25:51 UTC+12, [email protected]: > >> IMO, you shouldn't be doing this at all. For all devices, simply store >> the raw ifHCInOctets / if HCOutOctets (etc) in prometheus. Then when you >> want to graph it or alert on it, use a PromQL query using rate() or irate(). >> >> Storing the raw data keeps you closer to the source of truth, and makes >> it easier to debug, be able to observe counter resets, etc. You can also >> answer questions like "what was the average rate over 5 seconds? Over 30 >> seconds? Over 5 minutes?" using the same source data. >> >> If for some reason you end up with some specific queries that are too >> expensive to run, then you can use recording rules to generate the results >> or partial results. But remember there are other ways to improve query >> performance for dashboards, e.g. https://github.com/Comcast/trickster >> >> -- > You received this message because you are subscribed to the Google Groups > "Prometheus Users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/prometheus-users/c4b5f9fc-a40a-4198-88e0-2d3ca3ce1e33n%40googlegroups.com > <https://groups.google.com/d/msgid/prometheus-users/c4b5f9fc-a40a-4198-88e0-2d3ca3ce1e33n%40googlegroups.com?utm_medium=email&utm_source=footer> > . > -- You received this message because you are subscribed to the Google Groups "Prometheus Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-users/CABbyFmriN%3Dwm60fki_aFQvf4ix3prap84s0dGKgEvVrR3v2y3g%40mail.gmail.com.
