when sending 100 parallel http request the http response time is very high .. whereas the took time in for each query response is very low. How can i improve the http response time to ealstic search.
What is the size and specification of your cluster? Which Elasticsearch version are you using? How much data do you have in the cluster?
1 shard .. 50 mb data .. version 7.7.0.. jvm heap size 8gb
It sounds like it is a single node. How many CPU cores?
If I recall correctly the took time indicates how long it took to execute the query but this excludes any time the request is queued up waiting to be executed. Is your search queue filled up?
since my es instance is running on aws .. i should tell the cores of the instance?
What instance type are you using?
using aws elastic search service.
Yes, but which instance type is powering the cluster?