我为一个客户创建了一个查询，我一直在调整它，试图让它正好满足他们的需求。除了一些奇怪的计数外，我几乎把它设置好了。经过一个严肃的 FACEPALM 时刻后，我意识到 COUNT 和 SUM 中的 DISTINCT 正在抛弃我的数字，所以我删除了它们。
但是，现在查询似乎永远运行，而过去需要 2-3 分钟。我让它在一夜之间运行，它在 5 小时后崩溃并出现以下错误：
由于文件组“DEFAULT”中的磁盘空间不足，无法为数据库“TEMPDB”分配新页面。通过删除文件组中的对象、向文件组中添加其他文件或为文件组中的现有文件设置自动增长来创建必要的空间。
tempdb.mdf 为 16Gig，我有 7 个 tempdb_mssql_#.ndf 文件，每个文件为 16Gig，我的日志文件为 4Gig，非常高兴地完全填满了我的硬盘。
我重新启动了服务器并缩小了数据库，所以我又回到了原点，但我不明白为什么会这样。
这是导致问题的查询：
SELECT
DISTINCT ap.[SourcePartID] AS [Assembly Part ID],
p.[PART_X] AS [Assembly Part #],
p.[DESCR_X] AS [Assembly Part Description],
oa2.[Part Count],
oa2.[Total # of Parts],
([dbo].[fn_getFactoryStdCost](ap.[SourcePartID])) AS [Factory Std Cost],
oa2.[# of Docs],
oa2.[# of Software],
oa2.[# of Std Cost Parts],
oa2.[# of HR Devices],
oa2.[Sum HR Devices],
oa2.[# of 3rd Party Devices],
oa2.[Sum 3rd Party Devices],
oa2.[# of Robots],
oa2.[Sum of Robots],
oa2.[# of Buy Parts],
oa2.[# of Make Parts],
oa2.[# of Ref Parts],
bom.[Make/Buy]
INTO bomSummary
FROM AllPartsList ap
LEFT JOIN visuser.EN_PART p
ON p.[EN_Part_ID] = ap.[SourcePartID]
LEFT JOIN bomBreakdown bom
ON bom.[SourcePartID] = ap.[SourcePartID]
OUTER APPLY (
SELECT
[Part Count] = COUNT( DISTINCT IIF( bom.[Qty] = 0, null, bom.[Component Part #]) ),
[Total # of Parts] = SUM(bom.[Qty]),
[# of Std Cost Parts] = COUNT( IIF( ( [dbo].[fn_getFactoryStdCost]([ComponentPartID]) > 0 ), bom.[Component Part #], null) ),
[# of Docs] = COUNT( IIF( bom.[Commodity Code] IN ('009', '072', '073', '075', '079', '082'), bom.[Component Part #], null) ), -- Commodity Codes: 009, 072, 073, 075, 079, 082 : Commodity ID: 15, 84, 85, 87, 81, 92
[# of Software] = COUNT( IIF( bom.[Commodity Code] IN ('034'), bom.[Component Part #], null) ), -- Commodity Code 034 : Commodity ID: 28
[# of HR Devices] = COUNT( IIF( ( bom.[Commodity Code] IN ('002') AND [dbo].[fn_getFactoryStdCost]([ComponentPartID]) > 0 AND bom.[Qty] > 0), bom.[Component Part #], null) ), -- Commodity Code 002 : Commodity ID: 11
[Sum HR Devices] = SUM( IIF( ( bom.[Commodity Code] IN ('002') AND [dbo].[fn_getFactoryStdCost]([ComponentPartID]) > 0 AND bom.[Qty] > 0), bom.[Qty], null) ), -- Commodity Code 002 : Commodity ID: 11
[# of 3rd Party Devices] = COUNT( IIF( bom.[Commodity Code] IN ('007'), bom.[Component Part #], null) ), -- Commodity Code 007 : Commodity ID: 5
[Sum 3rd Party Devices] = SUM( IIF( ( bom.[Commodity Code] IN ('007') AND bom.[Qty] > 0), bom.[Qty], null) ), -- Commodity Code 007 : Commodity ID: 5
[# of Robots] = COUNT( IIF( ( bom.[Commodity Code] IN ('005') AND bom.[Make/Buy] = 'B' ), bom.[Component Part #], null) ), -- Commodity Code 005 : Commodity ID: 13
[Sum of Robots] = SUM( IIF( ( bom.[Commodity Code] IN ('005') AND bom.[Make/Buy] = 'B' ), bom.[Qty], null) ), -- Commodity Code 005 : Commodity ID: 13
[# of Make Parts] = COUNT( IIF( ( bom.[Make/Buy] = 'M' AND [dbo].[fn_getFactoryStdCost]([ComponentPartID]) > 0 AND bom.[Qty] > 0 ), bom.[Component Part #], null) ),
[# of Buy Parts] = COUNT( IIF( ( bom.[Make/Buy] = 'B' AND [dbo].[fn_getFactoryStdCost]([ComponentPartID]) > 0 AND bom.[Qty] > 0), bom.[Component Part #], null) ),
[# of Ref Parts] = COUNT( IIF( ( bom.[Make/Buy] = 'B' AND [dbo].[fn_getFactoryStdCost]([ComponentPartID]) = 0 ), bom.[Component Part #], null) )
FROM bomBreakdown bom
WHERE
bom.[ComponentPartID] IS NOT NULL AND
bom.[SourcePartID] = ap.[SourcePartID]
GROUP BY bom.[SourcePartID]
) oa2
ORDER BY p.[PART_X]
- AllPartsList 有 130 万条记录
- E_Part 有 20,500 条记录
- bomBreakdown 有 130 万条记录
以前，在外部应用中，除前两个之外的所有查询都有 DISTINCT，但我删除了它们。使用 DISTINCT 查询运行良好，数字刚刚关闭。
我不认为这是数据/表格布局问题，我认为可能存在交叉连接或我遗漏的东西。