Best Practices for MongoDB Indexes

Preface

Most developers know that indexing is faster. But in the actual process, we often encounter some questions & difficulties:

• The fields we query will have various cases. Do all fields involved in the query need to be indexed?
• How to choose between compound index and single field? Is it better to add both or a single field for each?
• Are there any side effects of adding index?
• The index has been added, but it’s still not fast enough? what to do?

This article attempts to explain the basic knowledge of indexing & answer the above questions.

1. What exactly is an index?

Most developers who come into contact with indexes probably know that indexes are similar to the catalog of books. You need to find the content you want, find the qualified keywords through the catalog, then find the pageno of the corresponding chapter, and then find the specific content. .
In the data structure, the simplest index implementation is similar to a hashmap, which maps to a specific location through the keyword key to find the specific content. But in addition to hashing, there are many ways to implement indexing.

(1) Multiple implementation methods and features of index

hash / b-tree / b -tree redis HSET / MongoDB&PostgreSQL / MySQL

hashmap

##See b-tree & b-tree difference in the picture:

b -tree leaves store data, non-leaves store indexes, no data is stored, and there are links between leaves
• b-tree non-leaves can store data

Algorithm search complexity:

hash is close to O(1)
• b-tree O(1)~ O(Log(n)) faster average search time , unstable query time
• b tree O(Log(n)) continuous data, query stability

As for why the implementation of MongoDB chooses b-tree instead of b - tree?

There are many articles on the Internet that explain this, but it is not the focus of this article.

(2) Storage of data & index

Index should be stored in memory as much as possible, followed by data. Be careful to keep only necessary indexes, and use memory as wisely as possible.
If the index memory is close to filling up the memory, it will be easy to read the disk and the speed will slow down.

(3) Thoughts after knowing the implementation & storage principle of index

insert/update/delete will trigger the rebalance tree, so if you add, delete or modify data, the index will trigger modifications, and performance will be lost. , the more indexes, the better. In this case, which fields should be selected as indexes? What should I do when the query uses these conditions?

Take the simplest hashmap as an example, why is the complexity not O(1), but so-called close to O(1). Because there are key conflicts/duplications, when the DB is looking for it, if there is a lot of data with key conflicts, it still has to take turns to continue looking. The same goes for b-tree looking at key selection.
So a mistake that most developers often make is to index keys that have no distinction. For example: many collections have only centralized categories of type/status documents with a count of hundreds of thousands or more. Usually this kind of index is not helpful.

2. Compound Index

(1) The more compound indexes, the better

If you don’t want to build more redundant indexes, the development colleagues will select compound & single fields. It's quite confusing sometimes. Let’s do a few experiments based on typical encounter scenarios:

A loans collection is created here. Simplified to only have 100 pieces of data. This is a loan table with _id, userId, status (loan status), amount (amount).

db.loans.count()100

db.loans.find({ "userId" : "59e022d33f239800129c61c7", "status" : "repayed", }).explain()
{
"queryPlanner" : {
"plannerVersion" : 1,
"namespace" : "cashLoan.loans",
"indexFilterSet" : false,
"parsedQuery" : {
 "$and" : [
   {
     "status" : {
       "$eq" : "repayed"
     }
   },
   {
     "userId" : {
       "$eq" : "59e022d33f239800129c61c7"
     }
   }
 ]
},
"queryHash" : "15D5A9A1",
"planCacheKey" : "15D5A9A1",
"winningPlan" : {
 "stage" : "COLLSCAN",
 "filter" : {
   "$and" : [
     {
       "status" : {
         "$eq" : "repayed"
       }
     },
     {
       "userId" : {
         "$eq" : "59e022d33f239800129c61c7"
       }
     }
   ]
 },
 "direction" : "forward"
},
"rejectedPlans" : [ ]
},
"serverInfo" : {
"host" : "RMBAP",
"port" : 27017,
"version" : "4.1.11",
"gitVersion" : "1b8a9f5dc5c3314042b55e7415a2a25045b32a94"
},
"ok" : 1
}

Note The COLLSCAN above scans the entire table because there is no index. Next we create several indexes respectively.

step 1 First create {userId:1, status:1}

db.loans.createIndex({userId:1, status:1})
{
"createdCollectionAutomatically" : false,
"numIndexesBefore" : 1,
"numIndexesAfter" : 2,
"ok" : 1
}

db.loans.find({ "userId" : "59e022d33f239800129c61c7", "status" : "repayed", }).explain()
{
"queryPlanner" : {
"plannerVersion" : 1,
"namespace" : "cashLoan.loans",
"indexFilterSet" : false,
"parsedQuery" : {
 "$and" : [
   {
     "status" : {
       "$eq" : "repayed"
     }
   },
   {
     "userId" : {
       "$eq" : "59e022d33f239800129c61c7"
     }
   }
 ]
},
"queryHash" : "15D5A9A1",
"planCacheKey" : "BB87F2BA",
"winningPlan" : {
 "stage" : "FETCH",
 "inputStage" : {
   "stage" : "IXSCAN",
   "keyPattern" : {
     "userId" : 1,
     "status" : 1
   },
   "indexName" : "userId_1_status_1",
   "isMultiKey" : false,
   "multiKeyPaths" : {
     "userId" : [ ],
     "status" : [ ]
   },
   "isUnique" : false,
   "isSparse" : false,
   "isPartial" : false,
   "indexVersion" : 2,
   "direction" : "forward",
   "indexBounds" : {
     "userId" : [
       "["59e022d33f239800129c61c7", "59e022d33f239800129c61c7"]"
     ],
     "status" : [
       "["repayed", "repayed"]"
     ]
   }
 }
},
"rejectedPlans" : [ ]
},
"serverInfo" : {
"host" : "RMBAP",
"port" : 27017,
"version" : "4.1.11",
"gitVersion" : "1b8a9f5dc5c3314042b55e7415a2a25045b32a94"
},
"ok" : 1
}

Result: {userId:1, status:1} is hit as the winning plan.

step2: Create a typical index userId

db.loans.createIndex({userId:1})
{
"createdCollectionAutomatically" : false,
"numIndexesBefore" : 2,
"numIndexesAfter" : 3,
"ok" : 1
}

db.loans.find({ "userId" : "59e022d33f239800129c61c7", "status" : "repayed", }).explain()
{
"queryPlanner" : {
"plannerVersion" : 1,
"namespace" : "cashLoan.loans",
"indexFilterSet" : false,
"parsedQuery" : {
 "$and" : [
   {
     "status" : {
       "$eq" : "repayed"
     }
   },
   {
     "userId" : {
       "$eq" : "59e022d33f239800129c61c7"
     }
   }
 ]
},
"queryHash" : "15D5A9A1",
"planCacheKey" : "1B1A4861",
"winningPlan" : {
 "stage" : "FETCH",
 "inputStage" : {
   "stage" : "IXSCAN",
   "keyPattern" : {
     "userId" : 1,
     "status" : 1
   },
   "indexName" : "userId_1_status_1",
   "isMultiKey" : false,
   "multiKeyPaths" : {
     "userId" : [ ],
     "status" : [ ]
   },
   "isUnique" : false,
   "isSparse" : false,
   "isPartial" : false,
   "indexVersion" : 2,
   "direction" : "forward",
   "indexBounds" : {
     "userId" : [
       "[\"59e022d33f239800129c61c7\", \"59e022d33f239800129c61c7\"]"
     ],
     "status" : [
       "[\"repayed\", \"repayed\"]"
     ]
   }
 }
},
"rejectedPlans" : [
 {
   "stage" : "FETCH",
   "filter" : {
     "status" : {
       "$eq" : "repayed"
     }
   },
   "inputStage" : {
     "stage" : "IXSCAN",
     "keyPattern" : {
       "userId" : 1
     },
     "indexName" : "userId_1",
     "isMultiKey" : false,
     "multiKeyPaths" : {
       "userId" : [ ]
     },
     "isUnique" : false,
     "isSparse" : false,
     "isPartial" : false,
     "indexVersion" : 2,
     "direction" : "forward",
     "indexBounds" : {
       "userId" : [
         "["59e022d33f239800129c61c7", "59e022d33f239800129c61c7"]"
       ]
     }
   }
 }
]
},
"serverInfo" : {
"host" : "RMBAP",
"port" : 27017,
"version" : "4.1.11",
"gitVersion" : "1b8a9f5dc5c3314042b55e7415a2a25045b32a94"
},
"ok" : 1
}

Note that DB detects {userId:1, status:1} as a better execution plan.

db.loans.find({ "userId" : "59e022d33f239800129c61c7" }).explain()
{
"queryPlanner" : {
"plannerVersion" : 1,
"namespace" : "cashLoan.loans",
"indexFilterSet" : false,
"parsedQuery" : {
 "userId" : {
   "$eq" : "59e022d33f239800129c61c7"
 }
},
"queryHash" : "B1777DBA",
"planCacheKey" : "1F09D68E",
"winningPlan" : {
 "stage" : "FETCH",
 "inputStage" : {
   "stage" : "IXSCAN",
   "keyPattern" : {
     "userId" : 1
   },
   "indexName" : "userId_1",
   "isMultiKey" : false,
   "multiKeyPaths" : {
     "userId" : [ ]
   },
   "isUnique" : false,
   "isSparse" : false,
   "isPartial" : false,
   "indexVersion" : 2,
   "direction" : "forward",
   "indexBounds" : {
     "userId" : [
       "["59e022d33f239800129c61c7", "59e022d33f239800129c61c7"]"
     ]
   }
 }
},
"rejectedPlans" : [
 {
   "stage" : "FETCH",
   "inputStage" : {
     "stage" : "IXSCAN",
     "keyPattern" : {
       "userId" : 1,
       "status" : 1
     },
     "indexName" : "userId_1_status_1",
     "isMultiKey" : false,
     "multiKeyPaths" : {
       "userId" : [ ],
       "status" : [ ]
     },
     "isUnique" : false,
     "isSparse" : false,
     "isPartial" : false,
     "indexVersion" : 2,
     "direction" : "forward",
     "indexBounds" : {
       "userId" : [
         "["59e022d33f239800129c61c7", "59e022d33f239800129c61c7"]"
       ],
       "status" : [
         "[MinKey, MaxKey]"
       ]
     }
   }
 }
]
},
"serverInfo" : {
"host" : "RMBAP",
"port" : 27017,
"version" : "4.1.11",
"gitVersion" : "1b8a9f5dc5c3314042b55e7415a2a25045b32a94"
},
"ok" : 1
}

Notice that DB detects {userId:1} as a better execution plan, um~, as we expected.

db.loans.find({ "status" : "repayed" }).explain()
{
"queryPlanner" : {
"plannerVersion" : 1,
"namespace" : "cashLoan.loans",
"indexFilterSet" : false,
"parsedQuery" : {
 "status" : {
   "$eq" : "repayed"
 }
},
"queryHash" : "E6304EB6",
"planCacheKey" : "7A94191B",
"winningPlan" : {
 "stage" : "COLLSCAN",
 "filter" : {
   "status" : {
     "$eq" : "repayed"
   }
 },
 "direction" : "forward"
},
"rejectedPlans" : [ ]
},
"serverInfo" : {
"host" : "RMBAP",
"port" : 27017,
"version" : "4.1.11",
"gitVersion" : "1b8a9f5dc5c3314042b55e7415a2a25045b32a94"
},
"ok" : 1
}

Interesting part: status does not hit the index, Full table scanNext step, add a sort:

db.loans.find({ "userId" : "59e022d33f239800129c61c7" }).sort({status:1}).explain()
{
"queryPlanner" : {
"plannerVersion" : 1,
"namespace" : "cashLoan.loans",
"indexFilterSet" : false,
"parsedQuery" : {
 "userId" : {
   "$eq" : "59e022d33f239800129c61c7"
 }
},
"queryHash" : "F5ABB1AA",
"planCacheKey" : "764CBAA8",
"winningPlan" : {
 "stage" : "FETCH",
 "inputStage" : {
   "stage" : "IXSCAN",
   "keyPattern" : {
     "userId" : 1,
     "status" : 1
   },
   "indexName" : "userId_1_status_1",
   "isMultiKey" : false,
   "multiKeyPaths" : {
     "userId" : [ ],
     "status" : [ ]
   },
   "isUnique" : false,
   "isSparse" : false,
   "isPartial" : false,
   "indexVersion" : 2,
   "direction" : "forward",
   "indexBounds" : {
     "userId" : [
       "["59e022d33f239800129c61c7", "59e022d33f239800129c61c7"]"
     ],
     "status" : [
       "[MinKey, MaxKey]"
     ]
   }
 }
},
"rejectedPlans" : [
 {
   "stage" : "SORT",
   "sortPattern" : {
     "status" : 1
   },
   "inputStage" : {
     "stage" : "SORT_KEY_GENERATOR",
     "inputStage" : {
       "stage" : "FETCH",
       "inputStage" : {
         "stage" : "IXSCAN",
         "keyPattern" : {
           "userId" : 1
         },
         "indexName" : "userId_1",
         "isMultiKey" : false,
         "multiKeyPaths" : {
           "userId" : [ ]
         },
         "isUnique" : false,
         "isSparse" : false,
         "isPartial" : false,
         "indexVersion" : 2,
         "direction" : "forward",
         "indexBounds" : {
           "userId" : [
             "["59e022d33f239800129c61c7", "59e022d33f239800129c61c7"]"
           ]
         }
       }
     }
   }
 }
]
},
"serverInfo" : {
"host" : "RMBAP",
"port" : 27017,
"version" : "4.1.11",
"gitVersion" : "1b8a9f5dc5c3314042b55e7415a2a25045b32a94"
},
"ok" : 1
}

(2) Other attempts

Interesting part: status does not hit the index

db.loans.find({ "status" : "repayed","userId" : "59e022d33f239800129c61c7", }).explain()
{
"queryPlanner" : {
"plannerVersion" : 1,
"namespace" : "cashLoan.loans",
"indexFilterSet" : false,
"parsedQuery" : {
 "$and" : [
   {
     "status" : {
       "$eq" : "repayed"
     }
   },
   {
     "userId" : {
       "$eq" : "59e022d33f239800129c61c7"
     }
   }
 ]
},
"queryHash" : "15D5A9A1",
"planCacheKey" : "1B1A4861",
"winningPlan" : {
 "stage" : "FETCH",
 "inputStage" : {
   "stage" : "IXSCAN",
   "keyPattern" : {
     "userId" : 1,
     "status" : 1
   },
   "indexName" : "userId_1_status_1",
   "isMultiKey" : false,
   "multiKeyPaths" : {
     "userId" : [ ],
     "status" : [ ]
   },
   "isUnique" : false,
   "isSparse" : false,
   "isPartial" : false,
   "indexVersion" : 2,
   "direction" : "forward",
   "indexBounds" : {
     "userId" : [
       "[\"59e022d33f239800129c61c7\", \"59e022d33f239800129c61c7\"]"
     ],
     "status" : [
       "[\"repayed\", \"repayed\"]"
     ]
   }
 }
},
"rejectedPlans" : [
 {
   "stage" : "FETCH",
   "filter" : {
     "status" : {
       "$eq" : "repayed"
     }
   },
   "inputStage" : {
     "stage" : "IXSCAN",
     "keyPattern" : {
       "userId" : 1
     },
     "indexName" : "userId_1",
     "isMultiKey" : false,
     "multiKeyPaths" : {
       "userId" : [ ]
     },
     "isUnique" : false,
     "isSparse" : false,
     "isPartial" : false,
     "indexVersion" : 2,
     "direction" : "forward",
     "indexBounds" : {
       "userId" : [
         "["59e022d33f239800129c61c7", "59e022d33f239800129c61c7"]"
       ]
     }
   }
 }
]
},
"serverInfo" : {
"host" : "RMBAP",
"port" : 27017,
"version" : "4.1.11",
"gitVersion" : "1b8a9f5dc5c3314042b55e7415a2a25045b32a94"
},
"ok" : 1
}

The hit index is not related to the order of each field of the query, as we guessed.

Come back to the interesting part, we delete the index {userId:1}

db.loans.dropIndex({"userId":1})
{ "nIndexesWas" : 3, "ok" : 1 }

db.loans.find({"userId" : "59e022d33f239800129c61c7", }).explain()
{
"queryPlanner" : {
"plannerVersion" : 1,
"namespace" : "cashLoan.loans",
"indexFilterSet" : false,
"parsedQuery" : {
 "userId" : {
   "$eq" : "59e022d33f239800129c61c7"
 }
},
"queryHash" : "B1777DBA",
"planCacheKey" : "5776AB9C",
"winningPlan" : {
 "stage" : "FETCH",
 "inputStage" : {
   "stage" : "IXSCAN",
   "keyPattern" : {
     "userId" : 1,
     "status" : 1
   },
   "indexName" : "userId_1_status_1",
   "isMultiKey" : false,
   "multiKeyPaths" : {
     "userId" : [ ],
     "status" : [ ]
   },
   "isUnique" : false,
   "isSparse" : false,
   "isPartial" : false,
   "indexVersion" : 2,
   "direction" : "forward",
   "indexBounds" : {
     "userId" : [
       "["59e022d33f239800129c61c7", "59e022d33f239800129c61c7"]"
     ],
     "status" : [
       "[MinKey, MaxKey]"
     ]
   }
 }
},
"rejectedPlans" : [ ]
},
"serverInfo" : {
"host" : "RMBAP",
"port" : 27017,
"version" : "4.1.11",
"gitVersion" : "1b8a9f5dc5c3314042b55e7415a2a25045b32a94"
},
"ok" : 1
}

DB execution analyzer thinks that the index {userId:1, status:1} can be better, but there is no hit Composite index, this is because status is not the leading field.

db.loans.find({ "status" : "repayed" }).explain()
{
"queryPlanner" : {
"plannerVersion" : 1,
"namespace" : "cashLoan.loans",
"indexFilterSet" : false,
"parsedQuery" : {
 "status" : {
   "$eq" : "repayed"
 }
},
"queryHash" : "E6304EB6",
"planCacheKey" : "7A94191B",
"winningPlan" : {
 "stage" : "COLLSCAN",
 "filter" : {
   "status" : {
     "$eq" : "repayed"
   }
 },
 "direction" : "forward"
},
"rejectedPlans" : [ ]
},
"serverInfo" : {
"host" : "RMBAP",
"port" : 27017,
"version" : "4.1.11",
"gitVersion" : "1b8a9f5dc5c3314042b55e7415a2a25045b32a94"
},
"ok" : 1
}

Change the sort angle again and interchange it with the previous query & sort. The previous one was:

db.loans.find({userId:1}).sort({ "status" : "repayed" })

See what’s the difference?

db.loans.find({ "status" : "repayed" }).sort({userId:1}).explain()
{
"queryPlanner" : {
"plannerVersion" : 1,
"namespace" : "cashLoan.loans",
"indexFilterSet" : false,
"parsedQuery" : {
 "status" : {
   "$eq" : "repayed"
 }
},
"queryHash" : "56EA6313",
"planCacheKey" : "2CFCDA7F",
"winningPlan" : {
 "stage" : "FETCH",
 "filter" : {
   "status" : {
     "$eq" : "repayed"
   }
 },
 "inputStage" : {
   "stage" : "IXSCAN",
   "keyPattern" : {
     "userId" : 1,
     "status" : 1
   },
   "indexName" : "userId_1_status_1",
   "isMultiKey" : false,
   "multiKeyPaths" : {
     "userId" : [ ],
     "status" : [ ]
   },
   "isUnique" : false,
   "isSparse" : false,
   "isPartial" : false,
   "indexVersion" : 2,
   "direction" : "forward",
   "indexBounds" : {
     "userId" : [
       "[MinKey, MaxKey]"
     ],
     "status" : [
       "[MinKey, MaxKey]"
     ]
   }
 }
},
"rejectedPlans" : [ ]
},
"serverInfo" : {
"host" : "RMBAP",
"port" : 27017,
"version" : "4.1.11",
"gitVersion" : "1b8a9f5dc5c3314042b55e7415a2a25045b32a94"
},
"ok" : 1
}

As guessed, hit the index.

Let’s play again and confirm the leading filed test:

db.loans.dropIndex("userId_1_status_1")
{ "nIndexesWas" : 2, "ok" : 1 }

db.loans.getIndexes()
[
{
"v" : 2,
"key" : {
 "id" : 1
},
"name" : "id_",
"ns" : "cashLoan.loans"
}
]

db.loans.createIndex({status:1, userId:1})
{
"createdCollectionAutomatically" : false,
"numIndexesBefore" : 1,
"numIndexesAfter" : 2,
"ok" : 1
}

db.loans.getIndexes()
[
{
"v" : 2,
"key" : {
 "id" : 1
},
"name" : "id_",
"ns" : "cashLoan.loans"
},
{
"v" : 2,
"key" : {
 "status" : 1,
 "userId" : 1
},
"name" : "status_1_userId_1",
"ns" : "cashLoan.loans"
}
]

db.loans.find({ "status" : "repayed" }).explain()
{
"queryPlanner" : {
"plannerVersion" : 1,
"namespace" : "cashLoan.loans",
"indexFilterSet" : false,
"parsedQuery" : {
 "status" : {
   "$eq" : "repayed"
 }
},
"queryHash" : "E6304EB6",
"planCacheKey" : "7A94191B",
"winningPlan" : {
 "stage" : "FETCH",
 "inputStage" : {
   "stage" : "IXSCAN",
   "keyPattern" : {
     "status" : 1,
     "userId" : 1
   },
   "indexName" : "status_1_userId_1",
   "isMultiKey" : false,
   "multiKeyPaths" : {
     "status" : [ ],
     "userId" : [ ]
   },
   "isUnique" : false,
   "isSparse" : false,
   "isPartial" : false,
   "indexVersion" : 2,
   "direction" : "forward",
   "indexBounds" : {
     "status" : [
       "["repayed", "repayed"]"
     ],
     "userId" : [
       "[MinKey, MaxKey]"
     ]
   }
 }
},
"rejectedPlans" : [ ]
},
"serverInfo" : {
"host" : "RMBAP",
"port" : 27017,
"version" : "4.1.11",
"gitVersion" : "1b8a9f5dc5c3314042b55e7415a2a25045b32a94"
},
"ok" : 1
}

db.loans.getIndexes()
[
{
"v" : 2,
"key" : {
 "id" : 1
},
"name" : "id_",
"ns" : "cashLoan.loans"
},
{
"v" : 2,
"key" : {
 "status" : 1,
 "userId" : 1
},
"name" : "status_1_userId_1",
"ns" : "cashLoan.loans"
}
]

db.loans.find({"userId" : "59e022d33f239800129c61c7", }).explain()
{
"queryPlanner" : {
"plannerVersion" : 1,
"namespace" : "cashLoan.loans",
"indexFilterSet" : false,
"parsedQuery" : {
 "userId" : {
   "$eq" : "59e022d33f239800129c61c7"
 }
},
"queryHash" : "B1777DBA",
"planCacheKey" : "5776AB9C",
"winningPlan" : {
 "stage" : "COLLSCAN",
 "filter" : {
   "userId" : {
     "$eq" : "59e022d33f239800129c61c7"
   }
 },
 "direction" : "forward"
},
"rejectedPlans" : [ ]
},
"serverInfo" : {
"host" : "RMBAP",
"port" : 27017,
"version" : "4.1.11",
"gitVersion" : "1b8a9f5dc5c3314042b55e7415a2a25045b32a94"
},
"ok" : 1
}

看完这个试验，明白了 {userId:1, status:1} vs {status:1,userId:1} 的差别了吗？

PS：这个case 里面其实status 区分度不高，这里只是作为实例展示。

三、总结：

• 注意使用上、使用频率上、区分高的/常用的在前面；
• 如果需要减少索引以节省memory/提高修改数据的性能的话，可以保留区分度高，常用的，去除区分度不高，不常用的索引。
• 学会用explain（）验证分析性能：

DB 一般都有执行器优化的分析，MySQL & MongoDB 都是 用explain 来做分析。
语法上MySQL :

explain your_sql

MongoDB:

yoursql.explain()

总结典型：理想的查询是结合explain 的指标，他们通常是多个的混合:

• IXSCAN  : 索引命中；
• Limit  : 带limit；
• Projection :  相当于非 select * ；
• Docs Size less is better  ；
• Docs Examined less is better ；
• nReturned=totalDocsExamined=totalKeysExamined ；
• SORT in index ：sort 也是命中索引，否则，需要拿到数据后，再执行一遍排序；
• Limit Array elements ： 限定数组返回的条数，数组也不应该太多数据，否则schema 设计不合理。

彩蛋

文末，还有最开头1个问题没回答：如果我的索引改加的都加了，还不够快,怎么办？
留个悬念，之后再写一篇。

更多PHP相关技术文章，请访问PHP教程栏目进行学习！

The above is the detailed content of Best Practices for MongoDB Indexes. For more information, please follow other related articles on the PHP Chinese website!