Elasticsearch read operations very high

issue: our cluster is observing high read operations

elasticsearch version: elasticsearch-6.2.1
5 nodes each with 2 cpus and 7GB or ram
the data volume for elasticsearch uses a zfs pool mounted using an ssd drive and the scheduler for that drive is set to noop

here is the node stats

here is the hot threads

i have also uploaded a diagram of the IO from a box and notice the avg reads per seconds is around 400.

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