I have a cluster for 12 data nodes running for a e commerce search, we also have 6 indices representing countries like US, CA, DE etc. I was running perf test against this cluster with just 1 index (i.e US - this comprises bulk of the data), the performance was in milliseconds. but when i add the remaining indices (most of the indices are kept as backup strategy for failover) performance degrades to seconds. can you please shed somelight on any configuration that assure performance in milliseconds though we have the back up indices on the cluster.
Hi @sara251186
Welcome to the community.
We need to understand a bit more on how you created these indices , how many shards and replicas you have , what is the size of index per shard , how much data is being written and how many queries per seconds etc.
but please go through these pages as the first step to start with:
1 Like
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.