"Limits to the number of shards which are searched in parallel"

The release notes for ES 6 mention:

Limits to the number of shards which are searched in parallel, so that a single query cannot dominate the cluster.

I haven't found documentation of how this feature works. I am wondering in particular whether it is possible to raise or remove these limits.

I am working on an internal application that has many nodes and few users. It has a query that is doing an aggregation using script that can take time to complete and I would love for it to be able to use all of the processor cores in the cluster.

I've come across action.search.shard_count.limit, but that seems to be different because that is a restriction on how many shards the query can touch at all, not how many it can touch at a time.

I found the answer. The setting is max_concurrent_shard_requests. It's documented here: https://www.elastic.co/guide/en/elasticsearch/reference/current/search-search.html , but I missed it because it appears to be part of the section "Multi-Index, Multi-Type”.

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