Is there any possibility to set weight for node resulting how many shards reside on this node? I mean eg I have 3 nodes - two big with more RAM and CPUs and one smaller with less RAM/CPU but with the same amount of disk space. I'd like this 3 node to be used less frequently and store less shards. I know about cluster wide settings allowing rerouting or per index setting where I can explicitly point to which nodes allocate shard but is there any node setting where I can achieve it automatically? BTW how much number of CPU cores / RAM size affects shards allocation. I couldn't find anything related in documentation.
Hi, will the third (lesser resourced) machine still be used for indexing, querying, as well as will it still reside replicas pertaining to the primaries found on the other two (beefier resourced) machines?
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.