如何提高mongodb的查询速度？

Question

我有一个使用mongodb的数据库，它有大约一百万个文档，并且还在不断增长。我用它来保持ping日志。随着它的增长，现在查询需要很长时间。

以下是我的查询（我正在使用pymongo）：

db['ping'].aggregate([
    {'$sort': {'url': 1, 'created': 1}},
    {'$group': {
        '_id': '$url',
        'lastCreated': {'$last': '$created'},
        'data': {'$first': '$$ROOT'}, },
     },
    {'$project': {
        'status': '$data.status',
        'date': '$lastCreated', }
     },
    {'$group': {
        '_id': '$status',
        'count': {'$sum': 1}, }
     }
], allowDiskUse=True)

此查询用于获取最新输入的数据，然后对其进行计数。

结果将如下所示：

{ "_id" : "CHECK", "count" : 8 }
{ "_id" : "DEAD", "count" : 7 }
{ "_id" : "OK", "count" : 94 }

以下是executionStats的说明：

{
    "stages" : [
        {
            "$cursor" : {
                "query" : {

                },
                "sort" : {
                    "url" : 1,
                    "created" : 1
                },
                "queryPlanner" : {
                    "plannerVersion" : 1,
                    "namespace" : "log.ping",
                    "indexFilterSet" : false,
                    "parsedQuery" : {

                    },
                    "winningPlan" : {
                        "stage" : "FETCH",
                        "inputStage" : {
                            "stage" : "IXSCAN",
                            "keyPattern" : {
                                "url" : 1,
                                "created" : 1
                            },
                            "indexName" : "url_1_created_1",
                            "isMultiKey" : false,
                            "multiKeyPaths" : {
                                "url" : [ ],
                                "created" : [ ]
                            },
                            "isUnique" : false,
                            "isSparse" : false,
                            "isPartial" : false,
                            "indexVersion" : 2,
                            "direction" : "forward",
                            "indexBounds" : {
                                "url" : [
                                    "[MinKey, MaxKey]"
                                ],
                                "created" : [
                                    "[MinKey, MaxKey]"
                                ]
                            }
                        }
                    },
                    "rejectedPlans" : [ ]
                },
                "executionStats" : {
                    "executionSuccess" : true,
                    "nReturned" : 1516493,
                    "executionTimeMillis" : 15105,
                    "totalKeysExamined" : 1516493,
                    "totalDocsExamined" : 1516493,
                    "executionStages" : {
                        "stage" : "FETCH",
                        "nReturned" : 1516493,
                        "executionTimeMillisEstimate" : 715,
                        "works" : 1516494,
                        "advanced" : 1516493,
                        "needTime" : 0,
                        "needYield" : 0,
                        "saveState" : 12104,
                        "restoreState" : 12104,
                        "isEOF" : 1,
                        "invalidates" : 0,
                        "docsExamined" : 1516493,
                        "alreadyHasObj" : 0,
                        "inputStage" : {
                            "stage" : "IXSCAN",
                            "nReturned" : 1516493,
                            "executionTimeMillisEstimate" : 268,
                            "works" : 1516494,
                            "advanced" : 1516493,
                            "needTime" : 0,
                            "needYield" : 0,
                            "saveState" : 12104,
                            "restoreState" : 12104,
                            "isEOF" : 1,
                            "invalidates" : 0,
                            "keyPattern" : {
                                "url" : 1,
                                "created" : 1
                            },
                            "indexName" : "url_1_created_1",
                            "isMultiKey" : false,
                            "multiKeyPaths" : {
                                "url" : [ ],
                                "created" : [ ]
                            },
                            "isUnique" : false,
                            "isSparse" : false,
                            "isPartial" : false,
                            "indexVersion" : 2,
                            "direction" : "forward",
                            "indexBounds" : {
                                "url" : [
                                    "[MinKey, MaxKey]"
                                ],
                                "created" : [
                                    "[MinKey, MaxKey]"
                                ]
                            },
                            "keysExamined" : 1516493,
                            "seeks" : 1,
                            "dupsTested" : 0,
                            "dupsDropped" : 0,
                            "seenInvalidated" : 0
                        }
                    }
                }
            }
        },
        {
            "$group" : {
                "_id" : "$url",
                "lastCreated" : {
                    "$last" : "$created"
                },
                "data" : {
                    "$first" : "$$ROOT"
                }
            }
        },
        {
            "$project" : {
                "_id" : true,
                "status" : "$data.status",
                "date" : "$lastCreated"
            }
        },
        {
            "$group" : {
                "_id" : "$status",
                "count" : {
                    "$sum" : {
                        "$const" : 1
                    }
                }
            }
        },
        {
            "$limit" : NumberLong(5)
        }
    ],
    "ok" : 1
}

我怀疑该查询检查了所有文档，导致它需要更长的时间。我确实尝试使用限制，但未对错误数据进行排序。

有没有办法使其运行更快？