PMM requirements for large setup

[mfedotov]

Hello,

We are investigating PMM as a monitoring system for a large installation (~500+ servers in each DC). And I wanted to ask, is it possible to handle such a big setup on a single PMM server and what capacity (CPU, MEM, HDD) it will require? Maybe you have already some experience with large setups and can point on problems and tuning solutions for them.

Thanks,
Max

[Peter]

Hi Max,

500 Nodes per server is on the higher side through we’ve seen this amount of instances work. It is possible with PMM 1.13 which uses Prometheus 2 which is much less resource intensive.

It is hard for me to give exact estimates but I would imagine pretty beefy box with no less than 16 cores and 32GB of memory. Amount of memory will be mostly driven by how many and what queries you run rather than data you’re streaming. For disk reasonably fast SSD should work.

If you have any load problems typically these are considerations to consider

1. Reduce resolution (ie METRICS_RESOLUTION) variable

2. Reduce amount of information you capture (for example table usage statistics or performance schema can generate a lot of data)

Let us know how it goes!

[Michael_Coburn]

Hi mfedotov
Just to add to Peter 's point - we have reports of a successful deployment of PMM against 1,400 MySQL instances. So, yes, this is on the higher side, and adjustments may be needed in terms of METRICS_RESOLUTION, but given 16 cores you should be able to reach 500 per DC when running PMM 1.13 or later.
​​​​​​​https://www.percona.com/doc/percona-monitoring-and-management/faq.html#what-resolution-is-used-for-metrics