mongodb查询语句效率分析
准备工作
- 安装最新版的mongodb.
- 安装免费的图形客户端Robo 3T
- 在数据库local下创建一个名为test的collection
- 创建索引
db.test.createIndex( { a: -1, b:-1 }, {name:"ab"}) - 准备一些数据
{_id:1, a:1, b:2}
{_id:2, a:1, b:2}
{_id:3, a:1, b:3}
{_id:4, a:1, b:3}
{_id:5, a:2, b:2}
{_id:6, a:2, b:2}
{_id:7, a:2, b:3}
{_id:8, a:2, b:3}
索引分析
db.test.find({a:1}).explain();
执行结果
{
"queryPlanner" : {
"plannerVersion" : 1,
"namespace" : "local.test",
"indexFilterSet" : false,
"parsedQuery" : {
"a" : {
"$eq" : 1.0
}
},
"winningPlan" : {
"stage" : "FETCH",
"inputStage" : {
"stage" : "IXSCAN",
"keyPattern" : {
"a" : -1.0,
"b" : -1.0
},
"indexName" : "ab",
"isMultiKey" : false,
"multiKeyPaths" : {
"a" : [],
"b" : []
},
"isUnique" : false,
"isSparse" : false,
"isPartial" : false,
"indexVersion" : 2,
"direction" : "forward",
"indexBounds" : {
"a" : [
"[1.0, 1.0]"
],
"b" : [
"[MaxKey, MinKey]"
]
}
}
},
"rejectedPlans" : []
},
"serverInfo" : {},
"ok" : 1.0
}
关键参数说明
| 字段 | 说明 |
|---|---|
| queryPlanner.indexFilterSet | 是否启用了index filter, mongodb使用index filter来选择索引 |
| queryPlanner.winningPlan | 优化器最终选择的plan |
| queryPlanner.winningPlan.inputStage | 查询输入的参数和索引 |
| queryPlanner.winningPlan.inputStage.stage | 所处的阶段COLLSCAN: 扫描collection IXSCAN扫描索引FETCH读取文档 |
| queryPlanner.winningPlan.inputStage.keyPattern | 索引的格式, 与创建索引时的格式一致 |
| queryPlanner.winningPlan.inputStage.indexBounds | 查询条件的约束值. |
| queryPlanner.rejectedPlans | 被拒绝掉的候选plan |
执行过程分析
db.test.find({a:1}).explain("executionStats");
{
"queryPlanner" : { 略 },
"executionStats" : {
"executionSuccess" : true,
"nReturned" : 4,
"executionTimeMillis" : 0,
"totalKeysExamined" : 4,
"totalDocsExamined" : 4,
"executionStages" : {
"stage" : "FETCH",
"nReturned" : 4,
"executionTimeMillisEstimate" : 0,
"works" : 5,
"advanced" : 4,
"needTime" : 0,
"needYield" : 0,
"saveState" : 0,
"restoreState" : 0,
"isEOF" : 1,
"invalidates" : 0,
"docsExamined" : 4,
"alreadyHasObj" : 0,
"inputStage" : {
"stage" : "IXSCAN",
"nReturned" : 4,
"executionTimeMillisEstimate" : 0,
"works" : 5,
"advanced" : 4,
"needTime" : 0,
"needYield" : 0,
"saveState" : 0,
"restoreState" : 0,
"isEOF" : 1,
"invalidates" : 0,
"keyPattern" : {
"a" : -1.0,
"b" : -1.0
},
"indexName" : "ab",
"isMultiKey" : false,
"multiKeyPaths" : {
"a" : [],
"b" : []
},
"isUnique" : false,
"isSparse" : false,
"isPartial" : false,
"indexVersion" : 2,
"direction" : "forward",
"indexBounds" : {
"a" : [
"[1.0, 1.0]"
],
"b" : [
"[MaxKey, MinKey]"
]
},
"keysExamined" : 4,
"seeks" : 1,
"dupsTested" : 0,
"dupsDropped" : 0,
"seenInvalidated" : 0
}
}
},
"serverInfo" : {},
"ok" : 1.0
}
关键字段说明
| 字段 | 说明 |
|---|---|
| executionStats.totalKeysExamined | 遍历索引的次数 |
| executionStats.totalDocsExamined | 遍历文档的次数 |
| executionTimeMillisEstimate | 预测需要执行的时间 |
遍历的次数越多, 遍历肯定越慢. 本例中使用了索引, 有4条满足条件的文档, 所有遍历次数都是4.
不使用索引
db.test.find({b:2}).explain("executionStats");
{
"executionStats" : {
"executionSuccess" : true,
"nReturned" : 4,
"executionTimeMillis" : 0,
"totalKeysExamined" : 0,
"totalDocsExamined" : 8,
},
}
db.test.find({b:2})无法使用索引{a:1, b:1}, 所以没有遍历索引, 然后遍历了整个collection
只返回索引数据
db.test.find({a:2}, {_id:0, a:1, b:1}).explain("executionStats");
{
"executionStats" : {
"executionSuccess" : true,
"nReturned" : 4,
"executionTimeMillis" : 0,
"totalKeysExamined" : 4,
"totalDocsExamined" : 0,
}
},
}
本例要mongodb只返回a, b两次字段, 而这两个字段刚好创建了索引, 不需要再读取document, 所以没有遍历文档.
总结
mongodb中的索引与mysql的索引的行为非常相似