es修改排序_【02】把 Elasticsearch 当数据库使：过滤和排序

最新推荐文章于 2024-04-23 10:26:13 发布
最新推荐文章于 2024-04-23 10:26:13 发布
阅读量409 收藏
点赞数
文章标签： es修改排序
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/weixin_30320081/article/details/111891801
版权
本文介绍了如何使用 Elasticsearch 进行复杂的查询和排序操作，包括使用 SQL 查询语法过滤文档，如 `exchange='nyse'` 和 `sector='Technology'`，以及按 `name` 字段进行升序排序。示例展示了如何利用工具 es_query.py 对 Elasticsearch 数据进行高效检索，包括布尔查询、范围查询、不等于查询等，并分析了查询性能。
摘要由CSDN通过智能技术生成

使用 https://github.com/taowen/es-monitor 可以用 SQL 进行 elasticsearch 的查询。本章介绍简单的文档过滤条件

exchange='nyse'

SQL

$ cat << EOF | ./es_query.py http://127.0.0.1:9200

select * from symbol where exchange='nyse' limit 1

EOF

{"sector": "n/a", "_type": "symbol", "market_cap": 4469064, "name": "3D Systems Corporation", "exchange": "nyse", "symbol": "DDD", "last_sale": 807, "_index": "symbol", "ipo_year": null, "_id": "AVLqbUjEQ3iIyyVFLQgV", "industry": "Technology"}

Elasticsearch

{

"query": {

"term": {

"exchange": "nyse"

}

},

"size": 1

}

{

"hits": {

"hits": [

{

"_score": 1.728313,

"_type": "symbol",

"_id": "AVLqbUjEQ3iIyyVFLQgV",

"_source": {

"sector": "n/a",

"market_cap": 4469064,

"name": "3D Systems Corporation",

"exchange": "nyse",

"symbol": "DDD",

"last_sale": 807,

"ipo_year": null,

"industry": "Technology"

},

"_index": "symbol"

}

],

"total": 3240,

"max_score": 1.728313

},

"_shards": {

"successful": 1,

"failed": 0,

"total": 1

},

"took": 6,

"timed_out": false

}

Profile

[

{

"query": [

{

"query_type": "TermQuery",

"lucene": "exchange:nyse",

"time": "2.336230000ms",

"breakdown": {

"score": 470159,

"create_weight": 417360,

"next_doc": 786055,

"match": 0,

"build_scorer": 662656,

"advance": 0

}

}

],

"rewrite_time": 3566,

"collector": [

{

"name": "SimpleTopScoreDocCollector",

"reason": "search_top_hits",

"time": "1.561495000ms"

}

]

}

]

exchange='nyse' AND sector='Technology'

SQL

$ cat << EOF | ./es_query.py http://127.0.0.1:9200

select * from symbol where exchange='nyse' and sector='Technology' limit 1

EOF

{"sector": "Technology", "_type": "symbol", "market_cap": 4469064, "name": "3D Systems Corporation", "exchange": "nyse", "symbol": "DDD", "last_sale": 807, "_index": "symbol", "ipo_year": null, "_id": "AVLqdtWhQ3iIyyVFLSJQ", "industry": "Computer Software: Prepackaged Software"}

Elasticsearch

{

"query": {

"bool": {

"filter": [

{

"term": {

"exchange": "nyse"

}

},

{

"term": {

"sector": "Technology"

}

}

]

}

},

"size": 1

}

{

"hits": {

"hits": [

{

"_score": 0.0,

"_type": "symbol",

"_id": "AVLqdtWhQ3iIyyVFLSJQ",

"_source": {

"sector": "Technology",

"market_cap": 4469064,

"name": "3D Systems Corporation",

"exchange": "nyse",

"symbol": "DDD",

"last_sale": 807,

"ipo_year": null,

"industry": "Computer Software: Prepackaged Software"

},

"_index": "symbol"

}

],

"total": 186,

"max_score": 0.0

},

"_shards": {

"successful": 1,

"failed": 0,

"total": 1

},

"took": 5,

"timed_out": false

}

Profile

[

{

"query": [

{

"query_type": "BooleanQuery",

"lucene": "#exchange:nyse #sector:Technology",

"time": "3.999703000ms",

"breakdown": {

"score": 54437,

"create_weight": 413531,

"next_doc": 1213355,

"match": 0,

"build_scorer": 736504,

"advance": 0

},

"children": [

{

"query_type": "TermQuery",

"lucene": "exchange:nyse",

"time": "1.175638000ms",

"breakdown": {

"score": 0,

"create_weight": 186745,

"next_doc": 0,

"match": 0,

"build_scorer": 616840,

"advance": 372053

}

},

{

"query_type": "TermQuery",

"lucene": "sector:Technology",

"time": "0.4062380000ms",

"breakdown": {

"score": 0,

"create_weight": 125321,

"next_doc": 115632,

"match": 0,

"build_scorer": 45136,

"advance": 120149

}

}

]

}

],

"rewrite_time": 23008,

"collector": [

{

"name": "SimpleTopScoreDocCollector",

"reason": "search_top_hits",

"time": "0.1031780000ms"

}

]

}

]

last_sale > 985

SQL

$ cat << EOF | ./es_query.py http://127.0.0.1:9200

select * from symbol where last_sale > 985 limit 1

EOF

{"sector": "Finance", "_type": "symbol", "market_cap": 42010000, "name": "1347 Capital Corp.", "exchange": "nasdaq", "symbol": "TFSCU", "last_sale": 1005, "_index": "symbol", "ipo_year": 2014, "_id": "AVLqdtWdQ3iIyyVFLRZB", "industry": "Business Services"}

Elasticsearch

{

"query": {

"range": {

"last_sale": {

"gt": 985.0

}

}

},

"size": 1

}

{

"hits": {

"hits": [

{

"_score": 1.0,

"_type": "symbol",

"_id": "AVLqdtWdQ3iIyyVFLRZB",

"_source": {

"sector": "Finance",

"market_cap": 42010000,

"name": "1347 Capital Corp.",

"exchange": "nasdaq",

"symbol": "TFSCU",

"last_sale": 1005,

"ipo_year": 2014,

"industry": "Business Services"

},

"_index": "symbol"

}

],

"total": 4285,

"max_score": 1.0

},

"_shards": {

"successful": 1,

"failed": 0,

"total": 1

},

"took": 2,

"timed_out": false

}

Profile

[

{

"query": [

{

"query_type": "MultiTermQueryConstantScoreWrapper",

"lucene": "last_sale:{985 TO *]",

"time": "4.684710000ms",

"breakdown": {

"score": 423966,

"create_weight": 21384,

"next_doc": 469032,

"match": 0,

"build_scorer": 3770328,

"advance": 0

}

}

],

"rewrite_time": 10493,

"collector": [

{

"name": "SimpleTopScoreDocCollector",

"reason": "search_top_hits",

"time": "1.261998000ms"

}

]

}

]

last_sale != 985

SQL

$ cat << EOF | ./es_query.py http://127.0.0.1:9200

select * from symbol where last_sale != 985 limit 1

EOF

{"sector": "Finance", "_type": "symbol", "market_cap": null, "name": "1347 Capital Corp.", "exchange": "nasdaq", "symbol": "TFSCR", "last_sale": 16, "_index": "symbol", "ipo_year": 2014, "_id": "AVLqdtWdQ3iIyyVFLRZA", "industry": "Business Services"}

Elasticsearch

{

"query": {

"bool": {

"must_not": {

"term": {

"last_sale": 985

}

}

}

},

"size": 1

}

{

"hits": {

"hits": [

{

"_score": 1.0,

"_type": "symbol",

"_id": "AVLqdtWdQ3iIyyVFLRZA",

"_source": {

"sector": "Finance",

"market_cap": null,

"name": "1347 Capital Corp.",

"exchange": "nasdaq",

"symbol": "TFSCR",

"last_sale": 16,

"ipo_year": 2014,

"industry": "Business Services"

},

"_index": "symbol"

}

],

"total": 6708,

"max_score": 1.0

},

"_shards": {

"successful": 1,

"failed": 0,

"total": 1

},

"took": 1,

"timed_out": false

}

Profile

[

{

"query": [

{

"query_type": "BooleanQuery",

"lucene": "+*:* -last_sale: Y",

"time": "14.10871500ms",

"breakdown": {

"score": 3397824,

"create_weight": 401619,

"next_doc": 7993655,

"match": 0,

"build_scorer": 141022,

"advance": 0

},

"children": [

{

"query_type": "MatchAllDocsQuery",

"lucene": "*:*",

"time": "1.694079000ms",

"breakdown": {

"score": 974401,

"create_weight": 5175,

"next_doc": 705676,

"match": 0,

"build_scorer": 8827,

"advance": 0

}

},

{

"query_type": "TermQuery",

"lucene": "last_sale: Y",

"time": "0.4805160000ms",

"breakdown": {

"score": 0,

"create_weight": 342905,

"next_doc": 0,

"match": 0,

"build_scorer": 104919,

"advance": 32692

}

}

]

}

],

"rewrite_time": 32619,

"collector": [

{

"name": "SimpleTopScoreDocCollector",

"reason": "search_top_hits",

"time": "4.604778000ms"

}

]

}

]

exchange='nyse' AND NOT sector='Technology'

SQL

$ cat << EOF | ./es_query.py http://127.0.0.1:9200

select * from symbol where exchange='nyse' AND NOT sector='Technology' limit 1

EOF

{"sector": "Health Care", "_type": "symbol", "market_cap": 3865437245, "name": "3M Company", "exchange": "nyse", "symbol": "MMM", "last_sale": 15244, "_index": "symbol", "ipo_year": null, "_id": "AVLqdtWhQ3iIyyVFLSJR", "industry": "Medical/Dental Instruments"}

Elasticsearch

{

"query": {

"bool": {

"filter": [

{

"term": {

"exchange": "nyse"

}

}

],

"must_not": [

{

"term": {

"sector": "Technology"

}

}

]

}

},

"size": 1

}

{

"hits": {

"hits": [

{

"_score": 0.0,

"_type": "symbol",

"_id": "AVLqdtWhQ3iIyyVFLSJR",

"_source": {

"sector": "Health Care",

"market_cap": 3865437245,

"name": "3M Company",

"exchange": "nyse",

"symbol": "MMM",

"last_sale": 15244,

"ipo_year": null,

"industry": "Medical/Dental Instruments"

},

"_index": "symbol"

}

],

"total": 3054,

"max_score": 0.0

},

"_shards": {

"successful": 1,

"failed": 0,

"total": 1

},

"took": 5,

"timed_out": false

}

Profile

[

{

"query": [

{

"query_type": "BooleanQuery",

"lucene": "#exchange:nyse -sector:Technology",

"time": "3.078346000ms",

"breakdown": {

"score": 164364,

"create_weight": 216094,

"next_doc": 1872817,

"match": 0,

"build_scorer": 114716,

"advance": 0

},

"children": [

{

"query_type": "TermQuery",

"lucene": "exchange:nyse",

"time": "0.4679310000ms",

"breakdown": {

"score": 0,

"create_weight": 119897,

"next_doc": 290703,

"match": 0,

"build_scorer": 57331,

"advance": 0

}

},

{

"query_type": "TermQuery",

"lucene": "sector:Technology",

"time": "0.2424240000ms",

"breakdown": {

"score": 0,

"create_weight": 62239,

"next_doc": 0,

"match": 0,

"build_scorer": 32978,

"advance": 147207

}

}

]

}

],

"rewrite_time": 19262,

"collector": [

{

"name": "SimpleTopScoreDocCollector",

"reason": "search_top_hits",

"time": "0.5360760000ms"

}

]

}

]

exchange='nyse' OR NOT sector='Technology'

SQL

$ cat << EOF | ./es_query.py http://127.0.0.1:9200

select * from symbol where exchange='nyse' OR NOT sector='Technology' limit 1

EOF

{"sector": "Health Care", "_type": "symbol", "market_cap": 3865437245, "name": "3M Company", "exchange": "nyse", "symbol": "MMM", "last_sale": 15244, "_index": "symbol", "ipo_year": null, "_id": "AVLqdtWhQ3iIyyVFLSJR", "industry": "Medical/Dental Instruments"}

Elasticsearch

{

"query": {

"bool": {

"should": [

{

"term": {

"exchange": "nyse"

}

},

{

"bool": {

"must_not": [

{

"term": {

"sector": "Technology"

}

}

]

}

}

]

}

},

"size": 1

}

{

"hits": {

"hits": [

{

"_score": 0.99838185,

"_type": "symbol",

"_id": "AVLqdtWhQ3iIyyVFLSJR",

"_source": {

"sector": "Health Care",

"market_cap": 3865437245,

"name": "3M Company",

"exchange": "nyse",

"symbol": "MMM",

"last_sale": 15244,

"ipo_year": null,

"industry": "Medical/Dental Instruments"

},

"_index": "symbol"

}

],

"total": 6245,

"max_score": 0.99838185

},

"_shards": {

"successful": 1,

"failed": 0,

"total": 1

},

"took": 9,

"timed_out": false

}

Profile

[

{

"query": [

{

"query_type": "BooleanQuery",

"lucene": "exchange:nyse (-sector:Technology +*:*)",

"time": "21.24846700ms",

"breakdown": {

"score": 4197895,

"create_weight": 815882,

"next_doc": 5963510,

"match": 0,

"build_scorer": 3902610,

"advance": 0

},

"children": [

{

"query_type": "TermQuery",

"lucene": "exchange:nyse",

"time": "1.099460000ms",

"breakdown": {

"score": 219612,

"create_weight": 529960,

"next_doc": 266196,

"match": 0,

"build_scorer": 83692,

"advance": 0

}

},

{

"query_type": "BooleanQuery",

"lucene": "-sector:Technology +*:*",

"time": "5.269110000ms",

"breakdown": {

"score": 1221522,

"create_weight": 209290,

"next_doc": 1693148,

"match": 0,

"build_scorer": 521166,

"advance": 0

},

"children": [

{

"query_type": "TermQuery",

"lucene": "sector:Technology",

"time": "0.6576720000ms",

"breakdown": {

"score": 0,

"create_weight": 141704,

"next_doc": 0,

"match": 0,

"build_scorer": 274371,

"advance": 241597

}

},

{

"query_type": "MatchAllDocsQuery",

"lucene": "*:*",

"time": "0.9663120000ms",

"breakdown": {

"score": 441766,

"create_weight": 5317,

"next_doc": 508525,

"match": 0,

"build_scorer": 10704,

"advance": 0

}

}

]

}

]

}

],

"rewrite_time": 61389,

"collector": [

{

"name": "SimpleTopScoreDocCollector",

"reason": "search_top_hits",

"time": "4.847758000ms"

}

]

}

]

date > now() - INTERVAL '60 DAYS'

SQL

$ cat << EOF | ./es_query.py http://127.0.0.1:9200

select * from quote where "date" > now() - interval '30 DAYS' limit 1

EOF

{"volume": 118900, "_type": "quote", "_index": "quote", "symbol": "AMRS", "adj_close": 147, "high": 149, "low": 138, "date": "2016-02-16", "close": 147, "_id": "AMRS-2016-02-16", "open": 140}

Elasticsearch

{

"query": {

"range": {

"date": {

"gt": 1453212988000

}

}

},

"size": 1

}

查询是和普通的大于一样的。

symbol LIKE 'AAP%'

SQL

$ cat << EOF | ./es_query.py http://127.0.0.1:9200

select * from symbol WHERE symbol LIKE 'AAP%'

EOF

{"sector": "Technology", "_type": "symbol", "market_cap": 522690000000, "name": "Apple Inc.", "exchange": "nasdaq", "symbol": "AAPL", "last_sale": 9427, "_index": "symbol", "ipo_year": 1980, "_id": "AVL0FqFJog4u0JP7lzka", "industry": "Computer Manufacturing"}

{"sector": "Finance", "_type": "symbol", "market_cap": 105960000, "name": "Atlantic Alliance Partnership Corp.", "exchange": "nasdaq", "symbol": "AAPC", "last_sale": 1019, "_index": "symbol", "ipo_year": 2015, "_id": "AVL0FqFJog4u0JP7lzlf", "industry": "Business Services"}

{"sector": "Consumer Services", "_type": "symbol", "market_cap": 148395246, "name": "Advance Auto Parts Inc", "exchange": "nyse", "symbol": "AAP", "last_sale": 13857, "_index": "symbol", "ipo_year": null, "_id": "AVL0FqFQog4u0JP7l0R7", "industry": "Other Specialty Stores"}

Elasticsearch

{

"query": {

"wildcard": {

"symbol": "AAP*"

}

}

}

{

"hits": {

"hits": [

{

"_score": 1.0,

"_type": "symbol",

"_id": "AVL0FqFJog4u0JP7lzka",

"_source": {

"sector": "Technology",

"market_cap": 522690000000,

"name": "Apple Inc.",

"exchange": "nasdaq",

"symbol": "AAPL",

"last_sale": 9427,

"ipo_year": 1980,

"industry": "Computer Manufacturing"

},

"_index": "symbol"

},

{

"_score": 1.0,

"_type": "symbol",

"_id": "AVL0FqFJog4u0JP7lzlf",

"_source": {

"sector": "Finance",

"market_cap": 105960000,

"name": "Atlantic Alliance Partnership Corp.",

"exchange": "nasdaq",

"symbol": "AAPC",

"last_sale": 1019,

"ipo_year": 2015,

"industry": "Business Services"

},

"_index": "symbol"

},

{

"_score": 1.0,

"_type": "symbol",

"_id": "AVL0FqFQog4u0JP7l0R7",

"_source": {

"sector": "Consumer Services",

"market_cap": 148395246,

"name": "Advance Auto Parts Inc",

"exchange": "nyse",

"symbol": "AAP",

"last_sale": 13857,

"ipo_year": null,

"industry": "Other Specialty Stores"

},

"_index": "symbol"

}

],

"total": 3,

"max_score": 1.0

},

"_shards": {

"successful": 1,

"failed": 0,

"total": 1

},

"took": 29,

"timed_out": false

}

Profile

[

{

"query": [

{

"query_type": "MultiTermQueryConstantScoreWrapper",

"lucene": "symbol:AAP*",

"time": "5.018644000ms",

"breakdown": {

"score": 3186,

"create_weight": 34308,

"next_doc": 31760,

"match": 0,

"build_scorer": 4949390,

"advance": 0

}

},

{

"query_type": "BooleanQuery",

"lucene": "symbol:AAP symbol:AAPC symbol:AAPL",

"time": "4.731495000ms",

"breakdown": {

"score": 0,

"create_weight": 205709,

"next_doc": 26459,

"match": 0,

"build_scorer": 4299185,

"advance": 0

},

"children": [

{

"query_type": "TermQuery",

"lucene": "symbol:AAP",

"time": "0.09350600000ms",

"breakdown": {

"score": 0,

"create_weight": 15337,

"next_doc": 4021,

"match": 0,

"build_scorer": 74148,

"advance": 0

}

},

{

"query_type": "TermQuery",

"lucene": "symbol:AAPC",

"time": "0.07045600000ms",

"breakdown": {

"score": 0,

"create_weight": 6874,

"next_doc": 1225,

"match": 0,

"build_scorer": 62357,

"advance": 0

}

},

{

"query_type": "TermQuery",

"lucene": "symbol:AAPL",

"time": "0.03618000000ms",

"breakdown": {

"score": 0,

"create_weight": 4934,

"next_doc": 1253,

"match": 0,

"build_scorer": 29993,

"advance": 0

}

}

]

}

],

"rewrite_time": 25703,

"collector": [

{

"name": "SimpleTopScoreDocCollector",

"reason": "search_top_hits",

"time": "0.04072600000ms"

}

]

}

]

}

]

从profile结果可以看出，所谓的LIKE查询，其实不是一个个去查文档。而是先从字典表里把这个字段的所有的符合LIKE条件的term查出来，然后去查满足这些term的文档。把LIKE变成一堆=条件的OR来处理的。

ORDER BY name

SQL

$ cat << EOF | ./es_query.py http://127.0.0.1:9200

select * from symbol order by name limit 1

EOF

{"sector": "n/a", "_type": "symbol", "market_cap": 72460000, "name": "iShares 0-5 Year Investment Grade Corporate Bond ETF", "exchange": "nasdaq", "symbol": "SLQD", "last_sale": 4997, "_index": "symbol", "ipo_year": null, "_id": "AVLqdtWfQ3iIyyVFLRwf", "industry": "n/a"}

Elasticsearch

{

"sort": [

{

"name": "asc"

}

],

"size": 1

}

{

"hits": {

"hits": [

{

"sort": [

"0"

],

"_type": "symbol",

"_source": {

"sector": "n/a",

"market_cap": 72460000,

"name": "iShares 0-5 Year Investment Grade Corporate Bond ETF",

"exchange": "nasdaq",

"symbol": "SLQD",

"last_sale": 4997,

"ipo_year": null,

"industry": "n/a"

},

"_score": null,

"_index": "symbol",

"_id": "AVLqdtWfQ3iIyyVFLRwf"

}

],

"total": 6714,

"max_score": null

},

"_shards": {

"successful": 1,

"failed": 0,

"total": 1

},

"took": 4,

"timed_out": false

}

Profile

[

{

"query": [

{

"query_type": "MatchAllDocsQuery",

"lucene": "*:*",

"time": "1.379354000ms",

"breakdown": {

"score": 0,

"create_weight": 32506,

"next_doc": 850386,

"match": 0,

"build_scorer": 496462,

"advance": 0

}

}

],

"rewrite_time": 18167,

"collector": [

{

"name": "SimpleFieldCollector",

"reason": "search_top_hits",

"time": "1.225077000ms"

}

]

}

]

weixin_30320081
关注
数据库面试题-ElasticSearch
Roc的博客
06-18 925
TF-IDF（Term Frequency - lnverse Document Frequency）是一种用于评估词语在文档集中的重要性的统计方法。它通过计算词频（TF）和逆文档频率（IDF）的乘积来工作。词频指一个词在文档中出现的次数，而逆文档频率是用来衡量一个词是否常见或是稀有的度量，计算方法是文档集中文档总数除以包含该词的文档数的对数。TF-IDF值越高，表示词在文档中越重要。
es修改排序_Elasticsearch - 根据名称进行过滤和按优先级排序
weixin_36121078的博客
01-13 1798
我想按名称筛选并按名称排序优先级异常，这意味着即使结果按字母顺序排序，我也希望首先显示特定名称。Elasticsearch - 根据名称进行过滤和按优先级排序例如 - 这是我的基本查询{"from": 0,"size": 500,"min_score": 0.15,"query": {"filtered": {"filter": {"bool": {"must": [{"exists": {"fi...
elasticsearch 7.15.3 实现内嵌字段排序 并且添加过滤排序条件
qq_39613976的博客
05-17 211
实现内嵌字段排序 并且添加过滤排序条件
深耕ElasticSearch - 过滤和聚合/多桶排序
你今天真好看呀
12-31 1762
文章目录1. 数据准备2. 过滤和聚合2.1 过滤2.2 过滤桶3. 多桶排序3.1 内置排序3.2 按度量排序4. 近似聚合 1. 数据准备 1、创建索引映射mapping ： PUT /cars { "mappings": { "properties": { "price":{ "type": "integer" }, "color":{ "type": "keyword" }, "make":{
es7.6.2 head 常用查询 排序 返回指定字段 高亮查询 指定多个字段检索 时间区间查询 now函数 wildcard查询 正则查询 多条件bool查询
boss_way的博客
09-14 1930
基本查询 排序 http://localhost:9200/suoyintest3/ PSOT _search { "query": { "match": { "content": "返回" } }, "sort": { "_score": "desc" } } 开启某个字段支持 fielddata 功能 http://localhost:9200/suoyintest3/_mapping/ put { "properties": ...
elasticsearch在评分相同的情况下排序
qq_19663899的博客
12-10 5830
写在前面 公司产品提出需求,需要对文档进行评分,如果评分不相同则按照评分大小排序,如果评分相同则在相同的结果下按照试题年份排序. 问题原因 es在执行sort之后方评分失效 解决办法 通过阅读文档得知,es在执行评分会大量耗时,并且在排序的时候由于没有指定元字段’_score’会导致不会进行评分.所以我们在使用的时候如果想根据评分相同的情况下在进行排序那我们就需要对评分进行选定. 选定评分的办法 ...
elasticsearch数据库下载、配置、使用案例PPT模板
最新发布
07-01
### Elasticsearch数据库的下载、配置与使用案例 #### 一、Elasticsearch数据库的下载与安装 **1.1 下载与解压** - **下载Elasticsearch：** - **官方网站下载：** 访问Elasticsearch官方网站...
Elasticsearch聚合的嵌套桶如何排序
zhendeC1的博客
04-23 750
针对前面提到的需求：统计每个汽车品牌下的每种颜色汽车的销售额，通常做法是：先按照品牌聚合，生成的每个桶(bucket)内有这个品牌的所有销售记录，然后将每个桶内的文档再按照颜色聚合，这样每个桶内就有多个子桶，每个子桶内就是每个品牌下每种颜色的销售记录。“make_color_sales_rank”: “desc” —排序字段是make_color_sales_rank，降序。对于这样的数据，无法做整体排序，因为内层桶的结果属于多值，而整体排序只能基于单值进行，以下是。
es查询大文本效率_Elasticsearch 技术分析（七）： Elasticsearch 的性能优化
weixin_39672296的博客
12-20 446
javajava8java开发Elasticsearch 技术分析(七)： Elasticsearch 的性能优化 硬件选择Elasticsearch(后文简称 ES)的基础是 Lucene，所有的索引和文档数据是存储在本地的磁盘中，具体的路径可在 ES 的配置文件../config/elasticsearch.yml中配置，如下：# --------------------...
es 排序 聚合统计_吐血整理：一文看懂ES的R，查询与聚合
weixin_39901439的博客
02-23 3327
对es查询的索引的company，其有如下字段，下面是一个示例数据"id": "1", //id"name": "张三",//姓名"sex": "男",//性别"age": 49,//年龄"birthday": "1970-01-01",//生日"position": "董事长",//职位"joinTime": "1990-01-01",//入职时间,日期格式"modified": "156216...
ES-使用profile做性能分析
wu892355227的博客
03-04 1941
本文简绍怎么使用profile做性能分析
ES学习笔记——ES入门：多条件查询、聚合、过滤、排序
Demonslzh的博客
11-12 1万+
文章目录一、es基础：轻量查询1、 增`PUT index_name/type_name/id`2、查（常用关键词）2.1 轻量查询:使用 ***_search***2.2 查询表达式2.2.1`match:匹配`2.2.2 `match_phrase:短语搜索`2.2.3 `range:过滤器，用于范围查询`二、ES聚合2.1`_bulk:批量添加数据`2.2 `_mapping操作`2.3 简单聚合2.4 添加度量指标聚合2.5 嵌套桶统计最大最小值2.6 按时间聚合统计每月汽车销量多条件时间聚合 一
es的range过滤_【ES从入门到实战】十五、全文检索-ElasticSearch-进阶-filter过滤
weixin_42506647的博客
01-14 518
接第14节6)、filter 【结果过滤】并不是所有的查询都需要产生分数，特别是那些仅用于 "fitering" (过滤)的文档。为了不计算分数 Elasticsearch 会自动检查场景并且优化查询的执行。在 filter 元素下指定的查询对得分没有影响-得分以 0 形式返回。分数仅受指定查询的影响。以 must 查询为例：GET /bank/_search{"query": {"bool": ...
Elasticsearch性能优化
Ghost Stories
02-08 1278
目录 批量提交 gateway 集群状态维护 缓存 字段数据 curator profiler 批量提交 在 CRUD 章节，我们已经知道 ES 的数据写入是如何操作的了。喜欢自己动手的读者可能已经迫不及待的自己写了程序开始往 ES 里写数据做测试。这时候大家会发现：程序的运行速度非常一般，即使 ES 服务运行在本机，一秒钟大概也就能写入几百条数据。 这种速度显然不...
ES5.6.4源码解析--聚合查询流程
qqqq0199181的博客
09-26 1390
aggregationPhase.preProcess(searchContext); 在执行聚合查询之前做一些准备：构建aggregationContext用于存放聚合过程中产生的数据；构建aggregators，他是聚合过程中统计数据的具体实现，由aggregator构成了采集器collectors。 ...
ES排序
weixin_38932035的博客
04-26 7702
排序 为了按照相关性来排序，需要将相关性表示为一个数值。在 Elasticsearch 中， 相关性得分 由一个浮点数进行表示，并在搜索结果中通过 _score 参数返回， 默认排序是 _score 降序。 有时，相关性评分对你来说并没有意义。例如，下面的查询返回所有 user_id 字段包含 1 的结果： GET /_search { "query" : { "bool...
【Elasticsearch】Elasticsearch 7.3 的 offheap 原理
九师兄
07-06 741
1.概述 转载：Elasticsearch 7.3 的 offheap 原理 一直以来，ES 堆中常驻内存中占据比重最大是 FST，即 tip(terms index) 文件占据的空间，1TB 索引大约占用2GB 或者更多的内存，因此为了节点稳定运行，业界通常认为一个节点 open 的索引不超过5TB。现在，从 ES 7.3版本开始，将 tip 文件修改为通过 mmap 的方式加载，这使 FST占据的内存从堆内转移到了堆外由操作系统的 pagecache 管理。 参考 ES 7.3 的 release-no
elasticsearch 深入 —— Top Hits Aggregation
热门推荐
gmHappy
09-19 2万+
Top Hits Aggregation top_hits指标聚合器跟踪正在聚合的最相关文档。 此聚合器旨在用作子聚合器，以便可以按桶聚合最匹配的文档。 top_hits聚合器可以有效地用于通过桶聚合器按特定字段对结果集进行分组。 一个或多个存储桶聚合器确定结果集被切入的属性。 选项 from - 要获取的第一个结果的偏移量。 size - 每个桶返回的最大匹配匹配数的最大数量。默认情况下...
Elasticsearch：top_hits aggregation
Elastic 中国社区官方博客
12-19 3055
top_hits指标聚合器跟踪要聚合的最相关文档。 该聚合器旨在用作子聚合器，以便可以按存储分区汇总最匹配的文档。 top_hits聚合器可以有效地用于通过存储桶聚合器按某些字段对结果集进行分组。 一个或多个存储桶聚合器确定将结果集切成哪些属性。 选项： from-要获取的第一个结果的偏移量。 size-每个存储桶要返回的最匹配匹配项的最大数目。 默认情况下，返回前三个匹配项。 排序-匹...
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值