We have a few year old cluster of 15 servers (2 ES nodes per server) with 8TBs of local storage each, but are we running low on space and having to cut back on retention. I want to add 80TBs to increase our total storage from 120TBs to 200TBs, but we are limited on physical rack space, so is it possible to add 4 servers with 20TBs of space each? (all numbers usable space after hardware RAID)
I've heard that Elasticsearch likes each ES node to be configured with the same amount of storage, is this true? I couldn't locate any documentation that explicitly states this. Would ES balance the shards ok across the node storage differences, or would 12 TBs be "wasted" on the new servers?
Unfortunately I don't think that would work for us. We have a lot of hot data, and we're not allowed to keep it past a certain amount of time, so we don't really have any cold data.
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.