INFORMATION
Why multiple nodes?
Primarily for resiliency: shards are replicated across all nodes, so if one node dies, elasticsearch automatically starts using the replica shards on the living nodes as primary shards (so there’s minimal downtime and data loss).
Our services will also automatically redirect any index request to the living nodes (once properly configured).
Additionally, multiple nodes help to provide faster search queries, because elasticsearch automatically balances the load between all nodes when searching: if a node is busy, elasticsearch will search using replica shards located on remote nodes (where CPU and memory is not under pressure).
MORE: For more information on determining how many nodes to use, and implementation please see the attached document.
APPLIES TO
Multimedia MiCC 8.0 and newer
Keywords: elasticsearch multi-node multi node multiple nodes