1. Basic Architecture of ElasticSearch

• when running ElasticSearch, we start up a node

• Node : instance of ElasticSearch that stores part of data

  • nodes can be distributed across multiple machines

• Cluster : collection of related nodes that together contain all of data

  • clusters are independent to each other
  • cross cluster search is possible, but not common
  • node either joins a cluster of create new cluster

2. Storing Data in ElasticSearch

• each unit of data stored within a cluster is called a document

• Documents : JSON objects

• when indexing a document, original json object is stored with added metadata

• Documents are organized within an index

  • when searching, index is specified so as to search within index

3. Acess Cluster

cluster exposes a REST API to communicate by

3-1. Kibana Console

ex1) check cluster health

  GET /_cluster/health

• _cluster : API, API begins with underscore
• health : Command

ex2) search

  GET /.kibana/_search {
      "query": {
          "match_all": {}
      }
  }

3-2. cURL

ex1) check cluster health

  $curl -xGET "http://localhost:9200/_cluster/health"

ex2) search

  $curl -xGET "http://localhost:9200/.kibana/_search" -H "Content-Type : application/json" -d"{ "query": { "match-all": {} } }"