Yes, this seems "on par" with what to expect from ~7M series. Typically I recommend sharding after 2M series per server, depending on the environment.
On Thu, Feb 27, 2020 at 10:33 AM Brian Candler <[email protected]> wrote: > 7 million timeseries is a very large number. Is it what you're > expecting? If not, then you should address this first. Maybe you have an > exporter which has a very high cardinality label which shouldn't be there. > You can get some clues from the prometheus GUI under Status > Runtime & > Build Information, including the highest cardinality labels and metrics. > > Try the RAM calculator which is on this page: > > https://www.robustperception.io/how-much-ram-does-prometheus-2-x-need-for-cardinality-and-ingestion > > 38GiB seems like the expected amount of RAM for what you're doing, *if* > you really need 7 million distinct timeseries. > > -- > 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/0ba46594-b7e5-4b16-b032-ec16a14ddc40%40googlegroups.com > <https://groups.google.com/d/msgid/prometheus-users/0ba46594-b7e5-4b16-b032-ec16a14ddc40%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/CABbyFmqDZwhCNHhYuW2gpyAe4s%3DduQ2061NYCpSgrN4BqG2jeA%40mail.gmail.com.
