docs: Resources for upgrades are not clear enough

[brunoluiz]

Exalate commented:

We are upgrading our cockroach cluster currently. There is a section in the 19.2 and 20.1 documentation which mention the following:

Make sure capacity and memory usage are reasonable for each node. Nodes must be able to tolerate some increase in case the new version uses more resources for your workload. Also go to Metrics > Dashboard: Hardware and make sure CPU percent is reasonable across the cluster. If there's not enough headroom on any of these metrics, consider adding nodes to your cluster before beginning your upgrade.

We were wondering, what would be "reasonable" for both capacity and memory usage in this case. It is not clear from the documentation what these values would be. For example one of our 19.2 clusters was with 70-80% memory usage pre-upgrade: is the 20~30% slack good enough for an upgrade?

reference 1 reference 2

Describe the solution you'd like

More clarity around what would be reasonable values for both capacity and memory usage pre-upgrade. If the value is "specific" to each cluster set-up or size, how could we calculate these values?

Describe alternatives you've considered

We could potentially increase all resources considerably for all nodes, to avoid potential issues during the upgrade. This wouldn't be optimal though.

Jira Issue: DOC-906

Epic DOC-11047