Question

我对SQL Server查询比较陌生。

我正在尝试编写一个查询，显示我公司的分支。我有3种方法可以这样做：

编写子查询：

Select
    (Select branchname 
        from branchtable 
        where branchPK = primarycompanybranchfk
    ) as branchone,
    (Select branchname 
        from branchtable 
        where branchPK = secondarycompanybranchfk
    ) as branchtwo
from companytable

使用联接：

Select 
    a.branchname as branchone,
    b.branchname as branchtwo
from companytable c
left join branchtable a     
    on a.branchPK = c.primarycompanybranchfk
left join branchtable b 
    on b.branchPK = c.secondarycompanybranchfk

使用子表连接：

 Select 
    branchfirst.branchname as branchone,
    branchsecond.branchname as branchtwo
 from companytable c
 left join (
    Select branchPK as prikey,
         branchname as branchname 
    from branchtable
 ) branchfirst 
        on branchfirst.prikey= c.primarycompanybranchfk
left join (
    Select branchPK as prikey,
        branchname as branchname 
    from branchtable
) branchsecond 
    on branchsecond.prikey= c.secondarycompanybranchfk

如果我使用第二个场景超过第二个场景，或者反之亦然，如果我的分支表包含超过10列，是否有任何性能提升？

Answer 1

还有一个提示:-)尝试设置每个人都可以重现的样本。使用自己的表会使这更难......

在大多数情况下，2和3将完全相同，＆＃34;嵌套列＆＃34;在这种情况下是最快的：

尝试此操作并设置＆＃34;包含执行计划＆＃34;选项。因此，您可以在之后比较执行计划（还有两个变体CROSS APPLY与2和3完全相同。）

究竟发生了什么 - 当然 - 在更复杂的情况下要复杂得多。这取决于现有的索引和统计数据。

SELECT c.COLUMN_NAME,
    (SELECT t.TABLE_NAME FROM INFORMATION_SCHEMA.TABLES AS t WHERE c.TABLE_NAME=t.TABLE_NAME) AS TABLE_NAME
from INFORMATION_SCHEMA.COLUMNS as c

SELECT c.COLUMN_NAME,t.TABLE_NAME
FROM INFORMATION_SCHEMA.COLUMNS as c
INNER JOIN INFORMATION_SCHEMA.TABLES AS t ON c.TABLE_NAME=t.TABLE_NAME

SELECT c.COLUMN_NAME,t.TABLE_NAME
FROM INFORMATION_SCHEMA.COLUMNS as c
INNER JOIN (SELECT t.TABLE_NAME FROM INFORMATION_SCHEMA.TABLES AS t) AS t ON c.TABLE_NAME=t.TABLE_NAME

SELECT c.COLUMN_NAME,t.TABLE_NAME
FROM INFORMATION_SCHEMA.COLUMNS as c
CROSS APPLY INFORMATION_SCHEMA.TABLES AS t 
WHERE c.TABLE_NAME=t.TABLE_NAME

SELECT c.COLUMN_NAME,t.TABLE_NAME
FROM INFORMATION_SCHEMA.COLUMNS as c
CROSS APPLY (SELECT t.TABLE_NAME FROM INFORMATION_SCHEMA.TABLES AS t WHERE c.TABLE_NAME=t.TABLE_NAME) AS t

Answer 2

在这种情况下，两者将具有相同的性能，因为查询优化器将两者视为相同的查询。但是如果你在嵌套查询中有一个过滤器，那么它会更快。

很好地加入所需的数据量而不是完整的数据集。

SQL Server：提高查询性能？

2 个答案: