需要帮助来提高MongoDB中查询的性能

Question

我正在使用MongoDB来实现一个POC（从SAP HANA迁移到MongoDB），因为我需要在查询中支持join，range和group by。这是我在RDBMS中的查询，我试图在MongoDB中复制它：

    SELECT *
FROM BUSINESSPOINTS_USA A
INNER JOIN (
    SELECT max(PKEY) AS PKEY
    FROM BUSINESSPOINTS_USA
    WHERE longitude BETWEEN % LL_LONG %
            AND % UR_LONG %
        AND latitude BETWEEN % LL_LAT %
            AND % UR_LAT %
        AND % WHERE_CLAUSE %
    GROUP BY LEFT(QUADKEY, % QUAD_TREE_LEVEL %)
    ) AS B ON A.PKEY = B.PKEY

但是group by在MongoDB中响应非常慢，这里是来自group by查询的部分：

db.getCollection('businesspoints').aggregate([{
                                            $match: {position : {$within: {$box: [[-73,-73],[41,41]]}}}},
                                                                {$group:{_id: {$substr:["$quadkey",0,5]}, mapinfoid:{$max: "$_id"}
                                                    }
                                           }]);

这是执行计划：

{
"waitedMS" : NumberLong(0),
"stages" : [ 
    {
        "$cursor" : {
            "query" : {
                "position" : {
                    "$within" : {
                        "$box" : [ 
                            [ 
                                -73.0, 
                                -73.0
                            ], 
                            [ 
                                41.0, 
                                41.0
                            ]
                        ]
                    }
                }
            },
            "fields" : {
                "quadkey" : 1,
                "_id" : 1
            },
            "queryPlanner" : {
                "plannerVersion" : 1,
                "namespace" : "spatialDB.businesspoints",
                "indexFilterSet" : false,
                "parsedQuery" : {
                    "position" : {
                        "$within" : {
                            "$box" : [ 
                                [ 
                                    -73.0, 
                                    -73.0
                                ], 
                                [ 
                                    41.0, 
                                    41.0
                                ]
                            ]
                        }
                    }
                },
                "winningPlan" : {
                    "stage" : "FETCH",
                    "filter" : {
                        "position" : {
                            "$within" : {
                                "$box" : [ 
                                    [ 
                                        -73.0, 
                                        -73.0
                                    ], 
                                    [ 
                                        41.0, 
                                        41.0
                                    ]
                                ]
                            }
                        }
                    },
                    "inputStage" : {
                        "stage" : "IXSCAN",
                        "keyPattern" : {
                            "position" : "2d"
                        },
                        "indexName" : "position_2d",
                        "isMultiKey" : false,
                        "isUnique" : false,
                        "isSparse" : false,
                        "isPartial" : false,
                        "indexVersion" : 1,
                        "direction" : "forward",
                        "indexBounds" : {
                            "position" : [ 
                                "[BinData(128, 30F0000000000000), BinData(128, 30FFFFFFFFFFFFFF)]", 
                                "[BinData(128, 3180000000000000), BinData(128, 31BFFFFFFFFFFFFF)]", 
                                "[BinData(128, 31C0000000000000), BinData(128, 31FFFFFFFFFFFFFF)]", 
                                "[BinData(128, 3200000000000000), BinData(128, 32FFFFFFFFFFFFFF)]", 
                                "[BinData(128, 3300000000000000), BinData(128, 33FFFFFFFFFFFFFF)]", 
                                "[BinData(128, 3400000000000000), BinData(128, 37FFFFFFFFFFFFFF)]", 
                                "[BinData(128, 3800000000000000), BinData(128, 3BFFFFFFFFFFFFFF)]", 
                                "[BinData(128, 3C00000000000000), BinData(128, 3FFFFFFFFFFFFFFF)]", 
                                "[BinData(128, 6000000000000000), BinData(128, 63FFFFFFFFFFFFFF)]", 
                                "[BinData(128, 6800000000000000), BinData(128, 6BFFFFFFFFFFFFFF)]", 
                                "[BinData(128, 9000000000000000), BinData(128, 93FFFFFFFFFFFFFF)]", 
                                "[BinData(128, 9400000000000000), BinData(128, 97FFFFFFFFFFFFFF)]", 
                                "[BinData(128, C000000000000000), BinData(128, C0FFFFFFFFFFFFFF)]", 
                                "[BinData(128, C100000000000000), BinData(128, C1FFFFFFFFFFFFFF)]", 
                                "[BinData(128, C200000000000000), BinData(128, C2FFFFFFFFFFFFFF)]", 
                                "[BinData(128, C300000000000000), BinData(128, C3FFFFFFFFFFFFFF)]"
                            ]
                        }
                    }
                },
                "rejectedPlans" : []
            }
        }
    }, 
    {
        "$group" : {
            "_id" : {
                "$substr" : [ 
                    "$quadkey", 
                    {
                        "$const" : 0.0
                    }, 
                    {
                        "$const" : 5.0
                    }
                ]
            },
            "mapinfoid" : {
                "$max" : "$_id"
            }
        }
    }
],
"ok" : 1.0
}

有没有办法将我的表分成多个集合来加速分组。