We have numa enabled server with 2 sockets, each socket with 28 cores and its one node cluster with 10 indies , 5 shards each.
Under the same workload(with high search operation load):
when we run ES7.3.0 on both the sockets , we can see CPU utilization is 30% , search throughput is 1700 ops/sec and when we increased the search load, the CPU utilization is not going beyond 30%. and throughput is also not going beyond 1700 ops/sec.
Is there any Elasticsearch specific settings which is restricting to use all available CPU's on high search load?
When we bind ES7.3.0 to one socket we can see all the CPU's were utilized around 90 to 95%.
and the ops/sec is 2550 ops/sec which is more then 1700 ops/sec.
Please let me know, in case of 2 sockets why the CPU is not scaling even with high search load.