如何获得数据库中所有用户创建的索引

Requirement

今天遇到一个SQL Performance Tuning需求：

对于用户表的索引进行优化，剔出那些重复的索引，添加必要的索引。

那么，我们必须先把表中的索引给找出来，SQL Server数据库和其他数据库一样，都有强大的元数据表，所以我们也可以从元数据表中得到索引(INDEX)的信息。 要完成这个查询需求，会涉及到master数据库中的几个视图(VIEW)：

• sys.[indexes]
• sys.[index_columns]
• sys.[all_columns]
• sys.[tables]
• sys.[schemas]

Resolution

SELECT '['+Sch.name+'].['+ Tab.[name]+']' AS TableName, 
       Ind.[name] AS IndexName, 
       SUBSTRING(( SELECT ', ' + AC.name 
                     FROM sys.[tables] AS T 
                    INNER JOIN sys.[indexes] I 
                       ON T.[object_id] = I.[object_id] 
                    INNER JOIN sys.[index_columns] IC 
                       ON I.[object_id] = IC.[object_id] 
                      AND I.[index_id] = IC.[index_id] 
                    INNER JOIN sys.[all_columns] AC 
                       ON T.[object_id] = AC.[object_id] 
                      AND IC.[column_id] = AC.[column_id] 
                    WHERE Ind.[object_id] = I.[object_id] 
                      AND Ind.index_id = I.index_id 
                      AND IC.is_included_column = 0 
                    ORDER BY IC.key_ordinal 
                      FOR XML PATH('') 
                 ), 2, 8000) AS KeyColumns, 
       SUBSTRING(( SELECT ', ' + AC.name 
                     FROM sys.[tables] AS T 
                    INNER JOIN sys.[indexes] I 
                       ON T.[object_id] = I.[object_id] 
                    INNER JOIN sys.[index_columns] IC 
                       ON I.[object_id] = IC.[object_id] 
                      AND I.[index_id] = IC.[index_id] 
                    INNER JOIN sys.[all_columns] AC 
                       ON T.[object_id] = AC.[object_id] 
                      AND IC.[column_id] = AC.[column_id] 
                    WHERE Ind.[object_id] = I.[object_id] 
                      AND Ind.index_id = I.index_id 
                      AND IC.is_included_column = 1 
                    ORDER BY IC.key_ordinal 
                      FOR XML PATH('') 
                 ), 2, 8000) AS IncludedColumns,
       Ind.type_desc AS IndexType,
       CASE Ind.is_primary_key
           WHEN 0 THEN 'N'
           WHEN 1 THEN 'Y'
       END AS IsPrimaryKey,
       CASE Ind.is_unique
           WHEN 0 THEN 'N'
           WHEN 1 THEN 'Y'
       END AS IsUnique
  FROM sys.[indexes] Ind 
 INNER JOIN sys.[tables] AS Tab 
    ON Tab.[object_id] = Ind.[object_id] 
 INNER JOIN sys.[schemas] AS Sch 
    ON Sch.[schema_id] = Tab.[schema_id] 
 WHERE Ind.[name] is not null
 ORDER BY TableName DESC

Explanation

上边这段代码的难点可能就是FOR XML PATH的用法。可以参考下面这篇博文：

灵活运用 SQL SERVER FOR XML PATH

Examples

在此我们拿Northwind数据库来演示代码执行后的结果。

TableName	IndexName	KeyColumns	IncludeColumns	IndexType	IsPrimaryKey	IsUnique
[dbo].[Territories]	PK_Territories	TerritoryID	NULL	NONCLUSTERED	Y	Y
[dbo].[Suppliers]	PK_Suppliers	SupplierID	NULL	CLUSTERED	Y	Y
[dbo].[Suppliers]	CompanyName	CompanyName	NULL	NONCLUSTERED	N	N
[dbo].[Suppliers]	PostalCode	PostalCode	NULL	NONCLUSTERED	N	N
[dbo].[Shippers]	PK_Shippers	ShipperID	NULL	CLUSTERED	Y	Y
[dbo].[Region]	PK_Region	RegionID	NULL	NONCLUSTERED	Y	Y
[dbo].[Products]	PK_Products	ProductID	NULL	CLUSTERED	Y	Y
[dbo].[Products]	CategoriesProducts	CategoryID	NULL	NONCLUSTERED	N	N