Proposal

Hi everyone, im trying to understand how much samples each of my recording/alerting rule is taking to understand how much memory i need for evaluating them. There are already metrics that shows how much time its taking to evaluate some alerting/recording rule groups. But ive encountered situation where applying too much rules led to high memory consumption and swapping. I think its important to know for better memory provisioning, how much memory will be needed for some rules before applying them.

Environment

• System information:

  Linux 3.10.0-693.17.1.el7.x86_64 x86_64

• Prometheus version:

  prometheus, version 2.9.1 (branch: HEAD, revision: ad71f27)
  build user: root@09f919068df4
  build date: 20190416-17:50:04
  go version: go1.12.4