如何加快对mongodb的查询？ 我将mongodriver repo用于.net核心

如何解决如何加快对mongodb的查询？ 我将mongodriver repo用于.net核心

我只查询特定日期范围内的 500.000 数据，但大约需要15-20分钟。

我使用了 linq ，但是结果是一样的。

使用回购链接：https://github.com/mongodb/mongo-csharp-driver

我的代码：

result = await collection
                               .Find(filter)
                               .Sort(sortBy)
                               .Skip(skip * limit)
                               .Limit(limit)
                               .ToListAsync();

编辑1：我们专注于服务器。希望问题解决。据我所知，通过服务器的访问速度很慢，但这与服务器功能无关，而与服务器之间的层有关。

编辑2：我的测试结果；

find({ "createDate" : { "$lte" : ISODate("2020-07-31T23:59:59.59Z"),"$gte" : ISODate("2020-07-01T00:00:00Z") } }).skip(0).limit(1000000).explain()

{
        "queryPlanner" : {
                "plannerVersion" : 1,"namespace" : "*","indexFilterSet" : false,"parsedQuery" : {
                        "$and" : [
                                {
                                        "createDate" : {
                                                "$lte" : ISODate("2020-07-31T23:59:59.590Z")
                                        }
                                },{
                                        "createDate" : {
                                                "$gte" : ISODate("2020-07-01T00:00:00Z")
                                        }
                                }
                        ]
                },"winningPlan" : {
                        "stage" : "LIMIT","limitAmount" : 1000000,"inputStage" : {
                                "stage" : "FETCH","inputStage" : {
                                        "stage" : "IXSCAN","keyPattern" : {
                                                "createDate" : 1
                                        },"indexName" : "createDate","isMultiKey" : false,"multiKeyPaths" : {
                                                "createDate" : [ ]
                                        },"isUnique" : false,"isSparse" : false,"isPartial" : false,"indexVersion" : 2,"direction" : "forward","indexBounds" : {
                                                "createDate" : [
                                                        "[new Date(1593561600000),new Date(1596239999590)]"
                                                ]
                                        }
                                }
                        }
                },"rejectedPlans" : [ ]
        },"serverInfo" : {
                "host" : "*","port" : *,"version" : "4.0.11","gitVersion" : "*"
        },"ok" : 1,"operationTime" : Timestamp(1597918530,1),"$clusterTime" : {
                "clusterTime" : Timestamp(1597918530,"signature" : {
                        "hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),"keyId" : NumberLong(0)
                }
        }
}

解决方法

加快数据库运行速度的最佳方法是从性能分析开始。您可以通过在数据库上执行以下操作来启用概要分析。

db.setProfilingLevel(2)

这将在数据库中创建一个system.profile集合，您可以在其中查询任何其他集合。

db.system.profile.find()

有关更多信息，请参见MongoDB docs。

一旦了解了很长时间，您就可以弄清楚哪些索引值得添加以支持正在执行的查询。

我还发现在C＃中的mongo客户端设置上设置“应用程序名称”属性很有用，这样您就可以查看查询来自何处。可以按以下步骤完成。

var client = new MongoClient(new MongoClientSettings
{
    ApplicationName = "StackoverflowExampleApp"
});

var db = client.GetDatabase("test");

var collection = db.GetCollection<BsonDocument>("test");

var findFluent = await collection.Find(FilterDefinition<BsonDocument>.Empty)
    .ToListAsync();

然后，如果我们设置了探查器并进行了查看。 我们将在个人资料文档中看到我们的appName

db.system.profile.find().pretty()
{
        "op" : "query","ns" : "test.test","command" : {
                "find" : "test","filter" : {

                },"$db" : "test","lsid" : {
                        "id" : UUID("9a49130c-7086-46c4-a564-a5ee4fbfc2a1")
                }
        },"keysExamined" : 0,"docsExamined" : 3,"cursorExhausted" : true,"numYield" : 0,"nreturned" : 3,"locks" : {
                "ReplicationStateTransition" : {
                        "acquireCount" : {
                                "w" : NumberLong(1)
                        }
                },"Global" : {
                        "acquireCount" : {
                                "r" : NumberLong(1)
                        }
                },"Database" : {
                        "acquireCount" : {
                                "r" : NumberLong(1)
                        }
                },"Collection" : {
                        "acquireCount" : {
                                "r" : NumberLong(1)
                        }
                },"Mutex" : {
                        "acquireCount" : {
                                "r" : NumberLong(1)
                        }
                }
        },"flowControl" : {

        },"responseLength" : 224,"protocol" : "op_msg","millis" : 2,"planSummary" : "COLLSCAN","execStats" : {
                "stage" : "COLLSCAN","nReturned" : 3,"executionTimeMillisEstimate" : 0,"works" : 5,"advanced" : 3,"needTime" : 1,"needYield" : 0,"saveState" : 0,"restoreState" : 0,"isEOF" : 1,"direction" : "forward","docsExamined" : 3
        },"ts" : ISODate("2020-08-14T09:20:11.323Z"),"client" : "127.0.0.1","appName" : "StackoverflowExampleApp","allUsers" : [ ],"user" : ""
}

，您将可以在日志中搜索为

db.system.profile.find({ appName: "StackoverflowExampleApp" }).pretty()