I have a Logstash instance sending data into an Elasticsearch cluster what take place on the same machine.
2 Elasticsearch and 1 Logstash.
Should i separate CPU Cores between them like this?
in case of 10 CPU:
logstash.yml: pipeline.workers: 5
elasticsearch.yml: node.processor: 5 (or just 2 or 3, because i have 2 Elasticsearch)
In this case is there a bare minimum requirement for CPU and RAM for 1-2.000 event per second, where the lines are 1-2 kb each?