Elasticsearch Disk Usage Internals

I'm trying to get a better understanding of elasticsearch internals - I've noticed when monitoring the disk usage of my elasticsearch cluster that there are periodic spikes in both master and data nodes during typical operation. This cluster has metric indices (streaming data) and other standard indices. Is this some sort of vacuuming process, or something else? I've attached a screenshot to show the spikes.

Screenshot from 2019-12-15 14-37-54

That could be explained when segment merge operations are happening on very big segments.

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