It is usually recommended to have the following index settings in case spinning disks are used for Elasticsearch data storage:
index.store.throttle.type: none
index.merge.scheduler.max_thread_count: 1
However, let's say a RAID0 array of 16 HDDs is used. Is the above recommendation still valid in that context, given that IO operations on RAID0 arrays are optimized?
Otherwise, how these settings could be changed to get the most out of the spinning HD RAID0-based storage in terms of performance?
