Elasticsearch Data Structure

Analyzes the Data Structure of Elasticsearch.

1. Elasticsearch Data Structure

[Figure 1] shows the Data Structure of Elasticsearch. Elasticsearch Data is composed of three levels: Index, Type, Document. Document stores Data in JSON format as a Tree structure. A collection of Documents is called Type. A collection of Types is called Index. Compared to MySQL, Index maps to Database, Type maps to Table, and Document maps to Row/Column.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15

{
  "-index" : "index",
  "-type" : "type",
  "-id" : "id",
  "-version" : 1,
  "-seq-no" : 0,
  "-primary-term" : 1,
  "found" : true,
  "-source" : {
    "key1" : "value1",
    "key2" : {
      "key3" : "value2"
    }
  }
}

[Data 1] shows information about a Document obtained from Elasticsearch. -index and -type represent the Index and Type to which the Document belongs. -id represents the ID of the Document. -version represents the Version of the Document, and the value of -version increases each time the Document is Updated. -source represents the actual Data of the Document.

2. References

• https://www.elastic.co/kr/blog/what-is-an-elasticsearch-index