inmemory Prometheus

[szuecs]

Proposal

As Prometheus user, that scrapes and queries only 1-5 minutes rate data, I would like to have the option to not save files and work fully inmemory, because the performance of Prometheus v2.2.1 in small Kubernetes clusters is not as I would expect.

Background

I deploy prometheus as statefulset with scrape interval 15s and have 1000m CPU and 4G memory set.
You can see the full configuration at https://github.com/zalando-incubator/kubernetes-on-aws/tree/dev/cluster/manifests/prometheus.

Monitor the monitor dashboard (Go metrics and Prometheus QPS and number of scrapes)

CPU metrics details

Memory metrics

[brian-brazil]

With such a setup, everything is already in memory. This sounds more like a usage question so it makes more sense to ask questions like this on the prometheus-users mailing list rather than in a GitHub issue. On the mailing list, more people are available to potentially respond to your question, and the whole community can benefit from the answers provided.

[szuecs]

https://groups.google.com/forum/#!topic/prometheus-users/2ECqkS0ZYKo

[lock]

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.