Apologies if I’ve missed the right documentation for this, but I’m looking for docs on how to balance/allocate overall total cluster memory.
What I’ve read indicates that each bucket gets some memory, and the cluster overall gets a total quota. But then there are services; I haven’t seen how to configure memory on those services individually.
For example, take this screenshot:
I have 2 buckets, 1GB total in the cluster; 1 bucket has 256MB, the other 128MB. Should bucket memory be allocated so the full 1GB is given to the buckets? If so then what do services use? How should this be budgeted? I’d like to give this test cluster for example 1GB total, but have it utilize as much of that as possible and not have any slack unallocated memory.