System requirements for M3DB cluster

[Boris Uzhinskiy]

Hi All.

Please help me to calculate CPU/MEM/Storage requirements for M3DB cluster.
We have a Prometheus with a load of 3k metrics  per second  and we need a HA decision for long-term storage (approx 1 year retention).

I plan to start with a 3 nodes in cluster and want to know a recomended system requirements to solve this.

Thank you

[Rob Skillington]

Hey Boris,

Sorry for the slow response, you can find us on Slack in the future if you'd like a faster response:

https://bit.ly/m3slack

If you're only doing 3k metrics/second you may want to think about continuing to use Prometheus with a large disk and setting the retention to 1 year running with two active Prometheus instance for HA. You can see how to configure that here (and set retention to "365d"):
https://prometheus.io/docs/prometheus/latest/storage/#operational-aspects

M3DB mainly focuses on providing horizontal scale out for when you're thinking that you'll eventually be collecting hundreds of thousands or millions of metrics/sec.