一、复合索引相关概述
1、复合索引创建语法
db.collection.createIndex( { <field1>: <type>, <field2>: <type2>, ... } )
同创建单键(列)索引一样,索引创建时需要指定每一个键索引的顺序
多个键直接用逗号分隔
索引创建语法可以参考:http://blog.youkuaiyun.com/leshami/article/details/53541978
2、复合索引的一些特性
复合索引可以支持要求匹配多个键的查询
复合索引每一个键的顺序非常重要,这将决定该索引在查询过程中能否被使用到
复合索引支持前导(缀)列索引查询
不能够创建基于哈希索引类型的复合索引
任意复合索引字段不能超过31个
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
二、复合索引示意图
如下图所示,在集合的userid以及score列上创建一个复合索引,其中userid为升序,score为降序
三、复合索引示例
1、演示环境
> db.version()
3.2.10
> db.example.find({},{"_id":0})
{ "id" : 1, "ename" : "leshami", "blog" : "http://blog.youkuaiyun.com/leshami", "name" : "leshami" }
演示集合数据,可以参考:http://blog.csdn.net/leshami/article/details/52672310
//查看任意的一个文档
> db.persons.find().limit(1).pretty()
{
"_id" : ObjectId("5812cbaaa129eed14b46458d"),
"name" : "robinson.cheng",
"age" : 25,
"email" : "robinson.cheng@qq.com",
"score" : {
"c" : 89,
"m" : 96,
"e" : 87
},
"country" : "USA",
"books" : [
"JS",
"C++",
"EXTJS",
"MONGODB"
]
}
- 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
2、创建复合索引
//如下示例,我们在集合persons上的name及age键上创建复合索引,且2个都为升序
> db.persons.createIndex({name:1,age:1})
{
"createdCollectionAutomatically" : false,
"numIndexesBefore" : 1,
"numIndexesAfter" : 2,
"ok" : 1
}
//在上面的示例中索引首先会按照name的值升序进行排列
//其次是age键,在name之后也按照升序排列
//下面过滤条件仅使用一个name键来查看执行计划
> db.persons.find({name:"robinson.cheng"}).explain()
{
......
"stage" : "FETCH",
"inputStage" : {
"stage" : "IXSCAN", //使用索引扫描
"keyPattern" : {
"name" : 1,
"age" : 1
},
"indexName" : "name_1_age_1",
......
"direction" : "forward",
"indexBounds" : {
"name" : [
"[\"robinson.cheng\", \"robinson.cheng\"]"
],
"age" : [
"[MinKey, MaxKey]"
.........
"ok" : 1
}
//下面过滤条件仅使用name及age键来查看执行计划
> db.persons.find({name:"robinson.cheng",age:25}).explain()
{
.........
"stage" : "FETCH",
"inputStage" : {
"stage" : "IXSCAN", //使用索引扫描
"keyPattern" : {
"name" : 1,
"age" : 1
},
"indexName" : "name_1_age_1",
.........
"direction" : "forward",
"indexBounds" : {
"name" : [
"[\"robinson.cheng\", \"robinson.cheng\"]"
],
"age" : [
"[25.0, 25.0]"
...........
"ok" : 1
}
//下面过滤条件仅使用name及age键来查看执行计划,但是将age键放在name键之前
> db.persons.find({age:25,name:"robinson.cheng"}).explain()
{
"queryPlanner" : {
.....
"winningPlan" : {
"stage" : "FETCH",
"inputStage" : {
"stage" : "IXSCAN", //使用索引扫描
"keyPattern" : {
"name" : 1,
"age" : 1
},
"indexName" : "name_1_age_1",
...........
"direction" : "forward",
"indexBounds" : {
"name" : [
"[\"robinson.cheng\", \"robinson.cheng\"]"
],
"age" : [
"[25.0, 25.0]"
........
"ok" : 1
}
//下面单独基于age键作为过滤条件进行查询
> db.persons.find({age:25}).explain()
{
................
"winningPlan" : {
"stage" : "COLLSCAN", //此处为使用集合扫描方式
"filter" : {
"age" : {
"$eq" : 25
}
},
"direction" : "forward"
},
"rejectedPlans" : [ ]
..............
"ok" : 1
}
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
- 13
- 14
- 15
- 16
- 17
- 18
- 19
- 20
- 21
- 22
- 23
- 24
- 25
- 26
- 27
- 28
- 29
- 30
- 31
- 32
- 33
- 34
- 35
- 36
- 37
- 38
- 39
- 40
- 41
- 42
- 43
- 44
- 45
- 46
- 47
- 48
- 49
- 50
- 51
- 52
- 53
- 54
- 55
- 56
- 57
- 58
- 59
- 60
- 61
- 62
- 63
- 64
- 65
- 66
- 67
- 68
- 69
- 70
- 71
- 72
- 73
- 74
- 75
- 76
- 77
- 78
- 79
- 80
- 81
- 82
- 83
- 84
- 85
- 86
- 87
- 88
- 89
- 90
- 91
- 92
- 93
- 94
- 95
- 96
- 97
- 98
- 99
- 100
- 101
- 102
- 103
3、复合索引与排序
复合索引创建时按升序或降序来指定其排列方式。对于单键索引,其顺序并不是特别重要,因为MongoDB可以在任一方向遍历索引
对于复合索引,按何种方式排序能够决定该索引在查询中能否被使用到。
//以下内容基于前面在{name:1,age:1}键上创建的索引来考察这个复合索引在排序时被使用到的场景
//基于{name:1,age:1}的排序
> db.persons.find().sort({name:1,age:1}).explain()
{
"queryPlanner" : {
....
"winningPlan" : {
"stage" : "FETCH",
"inputStage" : {
"stage" : "IXSCAN", //索引扫描
"keyPattern" : {
"name" : 1,
"age" : 1
},
"indexName" : "name_1_age_1",
..........
"direction" : "forward",
"indexBounds" : {
"name" : [
"[MinKey, MaxKey]"
],
"age" : [
"[MinKey, MaxKey]"
....
"ok" : 1
}
//基于{name:1,age:-1}的排序
> db.persons.find().sort({name:1,age:-1}).explain()
{
"queryPlanner" : {
....
"winningPlan" : {
"stage" : "SORT",
"sortPattern" : {
"name" : 1,
"age" : -1
},
"inputStage" : {
"stage" : "SORT_KEY_GENERATOR",
"inputStage" : {
"stage" : "COLLSCAN", //集合扫描
"filter" : {
"$and" : [ ]
},
"direction" : "forward"
.........
"ok" : 1
}
//基于{name:-1,age:1}的排序
> db.persons.find().sort({name:-1,age:1}).explain()
{
"queryPlanner" : {
.........
"winningPlan" : {
"stage" : "SORT",
"sortPattern" : {
"name" : -1,
"age" : 1
},
"inputStage" : {
"stage" : "SORT_KEY_GENERATOR",
"inputStage" : {
"stage" : "COLLSCAN", //集合扫描
"filter" : {
"$and" : [ ]
},
"direction" : "forward"
.........
"ok" : 1
}
//基于{age:1,name:1}的排序
> db.persons.find().sort({age:1,name:1}).explain()
{
....
"inputStage" : {
"stage" : "SORT_KEY_GENERATOR",
"inputStage" : {
"stage" : "COLLSCAN", //集合扫描
"filter" : {
"$and" : [ ]
},
"direction" : "forward"
..........
"ok" : 1
}
//基于{age:-1,name:1}的排序
> db.persons.find().sort({age:-1,name:1}).explain()
{
..........
"inputStage" : {
"stage" : "SORT_KEY_GENERATOR",
"inputStage" : {
"stage" : "COLLSCAN", //集合扫描
"filter" : {
"$and" : [ ]
},
"direction" : "forward"
..........
"ok" : 1
}
//基于{name:-1,age:-1}的排序
> db.persons.find().sort({name:-1,age:-1}).explain()
{
.........
"inputStage" : {
"stage" : "IXSCAN", //索引扫描
"keyPattern" : {
"name" : 1,
"age" : 1
},
"indexName" : "name_1_age_1",
............
"direction" : "backward", //注意,这里的方向为向后扫描
"indexBounds" : {
"name" : [
"[MaxKey, MinKey]"
],
"age" : [
"[MaxKey, MinKey]"
......
"ok" : 1
}
通过上面的不同场景,得出如下:
排序使用到索引的情形
db.persons.find().sort({name:1,age:1})
db.persons.find().sort({name:-1,age:-1})
排序未使用到索引的情形
db.persons.find().sort({name:1,age:-1})
db.persons.find().sort({name:-1,age:1})
db.persons.find().sort({age:1,name:1})
db.persons.find().sort({age:-1,name:1})
- 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
- 39
- 40
- 41
- 42
- 43
- 44
- 45
- 46
- 47
- 48
- 49
- 50
- 51
- 52
- 53
- 54
- 55
- 56
- 57
- 58
- 59
- 60
- 61
- 62
- 63
- 64
- 65
- 66
- 67
- 68
- 69
- 70
- 71
- 72
- 73
- 74
- 75
- 76
- 77
- 78
- 79
- 80
- 81
- 82
- 83
- 84
- 85
- 86
- 87
- 88
- 89
- 90
- 91
- 92
- 93
- 94
- 95
- 96
- 97
- 98
- 99
- 100
- 101
- 102
- 103
- 104
- 105
- 106
- 107
- 108
- 109
- 110
- 111
- 112
- 113
- 114
- 115
- 116
- 117
- 118
- 119
- 120
- 121
- 122
- 123
- 124
- 125
- 126
- 127
- 128
- 129
- 130
- 131
- 132
- 133
- 134
- 135
- 136
- 137
- 138
- 139
- 140
- 141
4、复合索引与索引前缀
索引前缀指的是复合索引的子集
假如存在如下索引
{ "item": 1, "location": 1, "stock": 1 }
那存在下列索引前缀
{ item: 1 }
{ item: 1, location: 1 }
在MongoDB中,下列查询过滤条件情形中,索引将会被使用到
item字段
item字段 + location字段
item字段 + location字段 + stock字段
item字段 + location字段(尽管索引被使用,但不高效)
以下过滤条件查询情形,索引将不会被使用到
location字段
stock字段
location + stock字段