Hi, I am running a solr cluster on AWS. And we chose instance types that can hold the whole lucene index size in memory. This was to avoid swapping.
I am now reviewing this choice and have a question to better understand solr RAM configuration. Let's say I have a lucene index size of 80GB and ONE solr node only. Can it be a consideration to have this solr node run on an instance of 32GB of RAM and a disc space > 200 GB? In this example, the query per sec is 200 and queries are unique (cache hit is low). Thanks, Moulay
