Distributed search performance

I currently have a 3 node cluster where each node has a complete copy of every index.

Based on the documentation on distributed search queries found here: https://www.elastic.co/guide/en/elasticsearch/guide/current/_query_phase.html

Do queries still go beyond the current node when searching if a node has all of the data it needs? If so, is there a way to configure things such that a query uses only the current node?

1 Like

You can control this by using preference, especially the _local one.

If, as he says, the index is fully replicated on every node in the cluster, won't ES automatically prefer the _local one?

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.