Scaling your clusters

While it's easy to get started with the launching of nodes and the forming of Elasticsearch clusters, the real challenge comes when the indexing and searching requests increase and your server encounters real pressure. In this section, we will discuss when and how to scale your Elasticsearch clusters.

Note

Cluster scaling is only possible if you have done some capacity planning in advance and have decided on an optimal number of shards. Always remember that once an index is created, you cannot increase or decrease the number of shards but can always change the number of replicas.

When to scale

Elasticsearch is very good at giving you hints in advance when it starts getting overloaded. The problems can arise in many areas such as slow searches, disk space utilizations, JVM issues, memory pressure, or high CPU utilizations. In any case, scaling should be done before your servers crash.

For timely scaling, the best thing to do is keep a close eye on the monitoring of the metrics...