白话Elasticsearch51-深入聚合数据分析之text field聚合以及fielddata原理

举报
小工匠 发表于 2021/09/11 00:07:22 2021/09/11
【摘要】 文章目录 概述官网示例对于分词的field执行aggregation,报错给分词的field,设置fielddata=true,可执行使用field.keyword,对分词的field进行聚合,可...


在这里插入图片描述


概述

继续跟中华石杉老师学习ES,第51篇

课程地址: https://www.roncoo.com/view/55


官网

fielddata: 戳这里

在这里插入图片描述
在这里插入图片描述

在这里插入图片描述

在这里插入图片描述


示例

对于分词的field执行aggregation,报错

先构造下模拟索引及数据

PUT /artisan_index 
{
  "mappings": {
    "artisan_type": {
      "properties": {
        "artisan_filed": {
          "type": "text"
        }
      }
    }
  }
}




PUT /artisan_index/artisan_type/1
{
  "artisan_filed": "artisan_1"
}

PUT /artisan_index/artisan_type/2
{
  "artisan_filed": "artisan_2"
}



  
 
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
GET  /artisan_index/_mapping/artisan_type

  
 
  • 1

在这里插入图片描述


我们建立的 artisan_filed是text类型,默认是分词的,那么我们对该字段进行 aggs看下

GET  /artisan_index/artisan_type/_search
{
  "size": 0,
  "aggs": {
    "group_by_artisan_field": {
      "terms": {
        "field": "artisan_filed"
      }
    }
  }
}


  
 
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12

报错如下:

 {
        "type": "illegal_argument_exception",
        "reason": "Fielddata is disabled on text fields by default. Set fielddata=true on [artisan_filed] in order to load fielddata in memory by uninverting the inverted index. Note that this can however use significant memory. Alternatively use a keyword field instead."
      }

  
 
  • 1
  • 2
  • 3
  • 4

在这里插入图片描述

对分词的field,直接执行聚合操作,会报错,大概意思是说,你必须要打开fielddata,然后将正排索引数据加载到内存中,才可以对分词的field执行聚合操作,而且会消耗很大的内存 .

当然了,排序这种操作也是不行的。

在这里插入图片描述


给分词的field,设置fielddata=true,可执行

#删除索引
DELETE artisan_index

#创建索引,设置text类型的字段的fielddata为true
PUT /artisan_index 
{
  "mappings": {
    "artisan_type": {
      "properties": {
        "artisan_filed": {
          "type": "text",
          "fielddata": true
        }
      }
    }
  }
}



#模拟数据
PUT /artisan_index/artisan_type/1
{
  "artisan_filed": "artisan_1"
}

PUT /artisan_index/artisan_type/2
{
  "artisan_filed": "artisan_2"
}


#查看映射
GET  /artisan_index/_mapping/artisan_type



---------------

  
 
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
  • 28
  • 29
  • 30
  • 31
  • 32
  • 33
  • 34
  • 35
  • 36
  • 37
  • 38

在这里插入图片描述

# 聚合操作
GET  /artisan_index/artisan_type/_search
{
  "size": 0,
  "aggs": {
    "group_by_artisan_field": {
      "terms": {
        "field": "artisan_filed"
      }
    }
  }
}




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

在这里插入图片描述

如果要对分词的field执行聚合操作,必须将fielddata设置为true


使用field.keyword,对分词的field进行聚合,可执行

#直接写入数据,让ES自动创建索引
PUT /artisan_index/artisan_type/1
{
  "artisan_filed": "artisan_1"
}

PUT /artisan_index/artisan_type/2
{
  "artisan_filed": "artisan_2"
}


#查看映射
GET  /artisan_index/_mapping/artisan_type



  
 
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16

在这里插入图片描述

# artisan_filed.keyword  es内置的keyword也可以在没设置fielddata=true的情况下聚合
GET  /artisan_index/artisan_type/_search
{
  "size": 0,
  "aggs": {
    "group_by_artisan_field": {
      "terms": {
        "field": "artisan_filed.keyword"
      }
    }
  }
}


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

artisan_filed.keyword es内置的keyword也可以在没设置fielddata=true的情况下。

当然了,如果对不分词的field执行聚合操作,直接就可以执行,不需要设置fieldata=true


分词field+fielddata的工作原理

doc value --> 不分词的所有field,可以执行聚合操作 --> 如果你的某个field不分词,那么在index-time,就会自动生成doc value --> 针对这些不分词的field执行聚合操作的时候,自动就会用doc value来执行。

分词field,是没有doc value的。。。在index-time,如果某个field是分词的,那么是不会给它建立doc value正排索引的,因为分词后,占用的空间过于大,所以默认是不支持分词field进行聚合的

分词field默认没有doc value,所以直接对分词field执行聚合操作,是会报错的

对于分词field,必须打开和使用fielddata,完全存在于纯内存中。。。结构和doc value类似。。。如果是ngram或者是大量term,那么必将占用大量的内存。。。

如果一定要对分词的field执行聚合,那么必须将fielddata=true,然后es就会在执行聚合操作的时候,现场将field对应的数据,建立一份fielddata正排索引,fielddata正排索引的结构跟doc value是类似的,但是只会将fielddata正排索引加载到内存中来,然后基于内存中的fielddata正排索引执行分词field的聚合操作

如果直接对分词field执行聚合,报错,提示让我们开启fielddata=true,告诉我们,会将fielddata uninverted index(正排索引),加载到内存,会耗费内存空间

为什么fielddata必须在内存?分词的字符串,需要按照term进行聚合,需要执行更加复杂的算法和操作,如果基于磁盘和os cache,那么性能会很差

文章来源: artisan.blog.csdn.net,作者:小小工匠,版权归原作者所有,如需转载,请联系作者。

原文链接:artisan.blog.csdn.net/article/details/100148048

【版权声明】本文为华为云社区用户转载文章,如果您发现本社区中有涉嫌抄袭的内容,欢迎发送邮件进行举报,并提供相关证据,一经查实,本社区将立刻删除涉嫌侵权内容,举报邮箱: cloudbbs@huaweicloud.com
  • 点赞
  • 收藏
  • 关注作者

评论(0

0/1000
抱歉,系统识别当前为高风险访问,暂不支持该操作

全部回复

上滑加载中

设置昵称

在此一键设置昵称,即可参与社区互动!

*长度不超过10个汉字或20个英文字符,设置后3个月内不可修改。

*长度不超过10个汉字或20个英文字符,设置后3个月内不可修改。