Scaling up elastic search using virtual CPUs (vCPUs)

Elastic Stack Elasticsearch
1

I am trying to index logdata using elastic search. I have been using a windows machine of 8gb ram. But now I want to scale it up and reduce the response time which comes in 10-20secs. So, I now want to run my elastic search on a virtual machine with 4vCPUs of 8gb ram each. The machine runs RHEL operating system. I tried the same indexing on this machine but got the same response time with no improvement. I think this is happening because elastic search is using only 1vCPU instead of all 4. As a result, I am getting almost similar performance. How can I make my elastic search to use all the vCPUs in order to improve the performance?

2

ES doesn't care whether your CPUs are physical or not. How many shards do your indexes have? What kind of queries are you making?

3

Hi thanks for the reply. I have 10 shards per index and I run queries similar to the following query-

	\"size\" : 0,
	\"aggs\" :
	{
		\"all-top-app_used\":
		{
			\"terms\": 
			{
				\"field\": \"AppName\",
					
				\"order\": 
				{
					\"sum11\": \"desc\"
					
				},
				\"size\" : 10
				
			},
			\"aggs\":
			{
				\"sum11\":
				{
					\"sum\":
					{
						\"field\": \"TotalTime\"
					}
				}
			}
		}
	}