此查询大约需要52秒才能完成。在订单表的表扫描中花费了82％的时间。它说实际的行数超过800万，所以我唯一的猜测就是它循环遍历这个表，而且我不熟悉SQL以了解原因。
订单表目前有19700行。
SELECT a.DistID, a.BusCtrID, d.FName, d.LName, r.Description
FROM funcGetDownline( 3 , 1) a
INNER JOIN Distributor d ON a.DistID = d.DistID
INNER JOIN Rank r ON d.RankID = r.RankID
INNER JOIN Orders o ON o.DistID = d.DistID
INNER JOIN orderlines ol ON ol.OrderID = o.OrderID
GROUP BY a.DistID, a.BusCtrID, d.FName, d.LName, r.Description'
funcGetDownline返回一个表，返回4,416条记录。如果我只对订单行执行查询，它将返回15,361行，这些行应该只返回不超过4,416行。
如果我对订单行执行查询，它将返回20811行，同样不应该返回超过4,416，少即可，但不是更多。它会在1秒内执行。
当我将其执行到组行时，执行此查询大约需要50秒，并返回4,313行，这似乎是正确的行数。
知道我在这里做错了什么吗？
再次编辑：
SELECT tmp.distid, tmp.busctrid, tmp.volume, r.description, d.fname, d.lname, d.email
FROM
(
SELECT o.Distid, o.busctrid, SUM (ol.volume * ol.quantity) as Volume
FROM Orders o
INNER JOIN Orderlines ol
ON ol.orderid = o.orderid
GROUP BY o.distid, o.busctrid
HAVING SUM (ol.volume * ol.quantity) BETWEEN 0 AND 3011
)tmp
INNER JOIN Distributor d
ON d.distid = tmp.distid
INNER JOIN Rank r
ON r.rankid = d.rankid
INNER JOIN FuncGetDownline(3,1) a
ON a.distid = tmp.distid
AND a.busctrid = tmp.busctrid
ORDER BY tmp.DistID, tmp.BusCtrID
上述查询在1秒或更短的时间内执行，我也不会为什么这么快。我的高级开发人员和我可能已经花了大约60分钟对其进行微调，以便快速进行调整。
EDITED LONG LIST：
orders PK_orders Orderid
orders PK_orders Distid
orders PK_orders BusCtrID
orders PK_orders ShipAmt
orders PK_orders ShipDate
orders PK_orders ShipTrack
orders PK_orders TaxAmt
orders PK_orders invoicetype
orders PK_orders Notes
orders PK_orders PostAmount
orders PK_orders Status
orders PK_orders ShippingLine1
orders PK_orders ShippingLine2
orders PK_orders ShippingCity
orders PK_orders ShippingState
orders PK_orders ShippingCountry
orders PK_orders ShippingPostalCode
orders PK_orders EnteredBy
orders PK_orders PostFuturePeriod
orders PK_orders OrderDate
orders PK_orders MO_Ship
orders PK_orders ts
orders PK_orders ShippingPhone
orders PK_orders DatePaid
orders PK_orders PublicNotes
orders PK_orders LastUpdatedBy
orders PK_orders BonusDate
orders PK_orders TypeCode
orders PK_orders ShippingName
orders PK_orders GeoCode
orders PK_orders InOutCity
orders PK_orders TaxOption
orders PK_orders IsCredit
orders PK_orders TaxAmt2
orders PK_orders Aristo_Status
orders PK_orders WarehouseID
orders PK_orders ShipCost
orders PK_orders Weight
orders PK_orders Carrier
orders PK_orders Service
orders PK_orders OrderExported
orders PK_orders ShippingEmail
orders PK_orders CreditOrderID
orders PK_orders ShipMethod
orders PK_orders OrderID2
orders PK_orders PartyID
orders PK_orders ExportTimeStamp
orders PK_orders OrderIP
orders PK_orders PromoCode
orders PK_orders FirstOrder
Distributor Dist_PK DistID
Distributor Dist_PK Name
Distributor Dist_PK Status
Distributor Dist_PK Rank
Distributor Dist_PK OverRank
Distributor Dist_PK ORankDate
Distributor Dist_PK EnterDate
Distributor Dist_PK RenewalDate
Distributor Dist_PK Password
Distributor Dist_PK RankID
Distributor Dist_PK LName
Distributor Dist_PK FName
Distributor Dist_PK EnteredBy
Distributor Dist_PK MInitial
Distributor Dist_PK Email
Distributor Dist_PK RankDate
Distributor Dist_PK SSN
Distributor Dist_PK URL
Distributor Dist_PK HomePhone
Distributor Dist_PK WorkPhone
Distributor Dist_PK Fax
Distributor Dist_PK BillLine1
Distributor Dist_PK BillLine2
Distributor Dist_PK BillCity
Distributor Dist_PK BillState
Distributor Dist_PK BillPostalCode
Distributor Dist_PK BillCountry
Distributor Dist_PK ShipLine1
Distributor Dist_PK ShipLine2
Distributor Dist_PK ShipCity
Distributor Dist_PK ShipState
Distributor Dist_PK ShipPostalCode
Distributor Dist_PK ShipCountry
Distributor Dist_PK ts
Distributor Dist_PK FirstCycle
Distributor Dist_PK DistFlag1
Distributor Dist_PK DistFlag2
Distributor Dist_PK DistFlag3
Distributor Dist_PK DistFlag4
Distributor Dist_PK DistFlag5
Distributor Dist_PK DistFlag6
Distributor Dist_PK DistFlag7
Distributor Dist_PK HomePhoneExt
Distributor Dist_PK WorkPhoneExt
Distributor Dist_PK FaxExt
Distributor Dist_PK ExtraNumeric1
Distributor Dist_PK ExtraNumeric2
Distributor Dist_PK ExtraNumeric3
Distributor Dist_PK ExtraSring1
Distributor Dist_PK ExtraSring2
Distributor Dist_PK ExtraSring3
Distributor Dist_PK TaxExempt
Distributor Dist_PK txjr
Distributor Dist_PK GeoCode
Distributor Dist_PK InOutCity
Distributor Dist_PK DefaultWarehouse
Distributor Dist_PK TaxAsCustomer
Distributor Dist_PK CellPhone
Distributor Dist_PK PaidAsRank
Distributor Dist_PK TaxID
Distributor Dist_PK BankID
Distributor Dist_PK TrainerID
Distributor Dist_PK Statement
Distributor Dist_PK WebShowName
Distributor Dist_PK WebAddress
Distributor Dist_PK WebHomePhone
Distributor Dist_PK WebWorkPhone
Distributor Dist_PK WebFax
Distributor Dist_PK WebUrl
Distributor Dist_PK WebEmail
Distributor Dist_PK Photo
Distributor Dist_PK WebTemplate
Distributor Dist_PK HTMLEmail
Distributor Dist_PK UserDef1
Distributor Dist_PK UserDef2
Distributor Dist_PK UserDef3
Distributor Dist_PK UserDef4
Distributor Dist_PK UserDef5
Distributor Dist_PK UserDef6
Distributor Dist_PK UserDef7
Distributor Dist_PK UserDef8
Distributor Dist_PK UserDef9
Distributor Dist_PK UserDef10
Distributor Dist_PK UserDef11
Distributor Dist_PK UserDef12
Distributor Dist_PK UserDef13
Distributor Dist_PK UserDef14
Distributor Dist_PK UserDef15
Distributor Dist_PK EnableEcommerce
Distributor Dist_PK BirthDate
Distributor Dist_PK ModifiedBy
Distributor Dist_PK UserName
Distributor Dist_PK LastUpdateDate
Distributor Dist_PK ReceiveCompanyEmail
Distributor Dist_PK ReceiveUplineEmail
Distributor Dist_PK ReceiveUplineMessages
Distributor Dist_PK UserDef16
Distributor Dist_PK UserDef17
Distributor Dist_PK UserDef18
Distributor Dist_PK UserDef19
Distributor Dist_PK UserDef20
Distributor Dist_PK CreatedIP
Distributor Dist_PK UserDef21
Distributor Dist_PK UserDef22
Distributor Dist_PK UserDef23
Distributor Dist_PK OverrideMinChkAmt
Distributor Dist_PK ReceiveSponsorEmail
Distributor Dist_PK ReceiveDownlineActivityEmail
Distributor Dist_PK DefaultLanguage
Distributor Dist_PK DistID2
Distributor Dist_PK ImHandle
Distributor Dist_PK IMType
Distributor Dist_PK ReceivePersonallySponsoredActivityEmail
Distributor Dist_PK EnableWebsite
Distributor Dist_PK DistributorCenterExpireDate
Distributor Dist_PK WebsiteExpireDate
Distributor Dist_PK AutologinSessionID
Distributor Dist_PK RecoverPasswordKey
Distributor Dist_PK PCIUserID
Distributor Dist_PK TLPCountry
Distributor Dist_PK EncryptedSSN
Distributor Dist_PK EncryptedTaxID
Distributor Dist_PK AgreementAccepted
orderlines PK_orderlines OrderID
orderlines PK_orderlines OrderLineID
orderlines PK_orderlines ItemId
orderlines PK_orderlines Quantity
orderlines PK_orderlines Ship
orderlines PK_orderlines Amount
orderlines PK_orderlines Tax
orderlines PK_orderlines Weight
orderlines PK_orderlines StatusDate
orderlines PK_orderlines ShipDate
orderlines PK_orderlines Status
orderlines PK_orderlines Tracking1
orderlines PK_orderlines Tracking2
orderlines PK_orderlines RetailPrice
orderlines PK_orderlines WholesalePrice
orderlines PK_orderlines GroupItem
orderlines PK_orderlines Volume
orderlines PK_orderlines ts
orderlines PK_orderlines WarehouseID
orderlines PK_orderlines Notes
orderlines PK_orderlines ExtraCurrency1
orderlines PK_orderlines ExtraCurrency2
orderlines PK_orderlines ExtraCurrency3
orderlines PK_orderlines TaxRecID
orderlines PK_orderlines TrackingShipCo
orderlines PK_orderlines TaxRate
orderlines PK_orderlines MaxTaxAmt
orderlines PK_orderlines MaxOrderAmt
orderlines PK_orderlines CustomPrice
orderlines PK_orderlines DiscountPrice
orderlines PK_orderlines DiscountId
orderlines PK_orderlines DiscountDescription
orderlines PK_orderlines ShipCost
orderlines PK_orderlines Aristo_Status
orderlines PK_orderlines First_Time_Orderitem
orderlines PK_orderlines AttributeDetail
orderlines PK_orderlines Description
orderlines PK_orderlines BoxNumber
答案 0 :(得分：3)
一些建议：
希望这有帮助。
答案 1 :(得分：3)
记录数量增加的原因是内部联接返回每个匹配的记录。
IE。
如果订单中有1个订单，订单行中有5个匹配的订单ID，则会返回5行。
所以你可能在一个点上有4416行，由于连接而很容易爆炸到更高的数字，然后由于你得到了你想要的数字
修改
Try this query.
SELECT
T.Name as TableName, I.name as IndexName,s.name as IndexColumn
FROM
sys.indexes I
INNER JOIN
Sys.tables t
ON
t.object_id = I.Object_ID
inner join
Sys.columns s
on
t.object_id = s.object_id
WHERE
index_id >= 1
and
t.type = 'U'
这应该让我们知道您对索引的位置。
您也可以发布生成的执行计划吗？如果从SQL Server Management Studio运行查询，请右键单击查询编写区域并选择包含实际执行计划。
答案 2 :(得分：1)
要检查的第一件事是你在join子句中的所有列都有索引。
第二，你的SQL表示分组，因此在分组之前20K的数字可能是正确的。
但是，group by子句通常表示组中的某种类型的数学，例如sum（）函数，我在您的sql中没有看到。您可以尝试使用distinct而不是group：
SELECT DISTINCT a.DistID, a.BusCtrID, d.FName, d.LName, r.Description
FROM funcGetDownline( 3 , 1) a INNER JOIN Distributor d ON a.DistID =
d.DistID INNER JOIN Rank r ON d.RankID = r.RankID INNER JOIN Orders o
ON o.DistID = d.DistID INNER JOIN orderlines ol ON ol.OrderID =
o.OrderID
答案 3 :(得分：1)
我不确定这一点，但看起来您不需要来自订单或订单行的数据，只是知道它们存在于此查询中，所以......
SELECT a.DistID, a.BusCtrID, d.FName, d.LName, r.Description
FROM funcGetDownline( 3 , 1) a
INNER JOIN Distributor d ON a.DistID = d.DistID
INNER JOIN Rank r ON d.RankID = r.RankID
where exists (select o.distid from Orders o
INNER JOIN orderlines ol ON ol.OrderID = o.OrderID
where o.DistID = d.DistID )