Does, number of indices has considerable affect on running/startup performance? like image having 100GB of data in 50 indices or having 100GB data on 300+ indices, how they differ from performance perspective? imagine I already has 100GB+ with 300+ indices does it worth to go and combine indices in more common groups to reduce number of indices?
1 Like
Shards matter, not indices.
While Elasticsearch 8.X is far more efficient in managing shards, you still don't want a tonne of really small ones.
Indices do matter for some performance aspects too - e.g. authorisation happens at the index level. But I would not expect there to be much difference between 50 indices and 300 indices, these are both fairly small numbers in modern Elasticsearch. Still, the answer really does depend on your specific setup, so the only way to be sure is to compare both approaches in a benchmark with a realistic workload. Fortunately, there's a tool to do exactly that: Rally.
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.