I have a ES 7.9.2, [ 10node cluster each: 4Core, 32G RAM, 500GB SSD]
I tried performance test on index , 5 primary shards, 0 replica.
It reaches around 200K events/sec
and I found weird performance drops. Can you help me how to troubleshoot performance drops.
Where should I look to determine the potential cause of the issue?
If you have 5 primary shards and no replica shards, only 5 data nodes will be processing indexing requests for this index. The periodicity in performance drops could very well be due to merging. Are you monitoring disk I/O and iowait on the data nodes? Is there anything in the Elasticsearch logs around these dips?
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.