Nodes

Elasticsearch is a distributed system. It consists of multiple processes running across different machines in a network that communicate with the other processes. In Chapter 1, Introducing the Elastic Stack, we downloaded, installed, and started Elasticsearch. It started what is called a single node of Elasticsearch, or a single node Elasticsearch cluster.

An Elasticsearch node is a single server of Elasticsearch, which may be part of a larger cluster of nodes. It participates in indexing, searching, and performing other operations that are supported by Elasticsearch. Every Elasticsearch node is assigned a unique ID and name when it is started. A node can also be assigned a static name via the node.name parameter in the Elasticsearch configuration file, config/elasticsearch.yml.

Every Elasticsearch node or instance has a main configuration file, which is located in the config subdirectory. The file is in YML format (which stands for YAML Ain't Markup Language). This configuration file can be used to change defaults such as the node name, port, and cluster name.

At the lowest level, a node corresponds to one instance of the Elasticsearch process. It is responsible for managing its share of data.