SqlServer性能优化 通过压缩与计算列提高性能(十一)

 压缩:

   1.压缩的对象

        1.表   2.索引(非聚集索引手工做)   3.备份(手工做)

   2.对性能影响

        1.提高IO性能     2.降低CPU性能

 行压缩:

     1.对null值不占用空间

     2.对Numeric值不占用空间

页压缩:

     1.行压缩

     2.前缀压缩

     3.字典压缩

   实例:

准备表数据:

select * from SalesOrderDetail

select * into ComOrderDetail from SalesOrderDetail

sp_spaceused 'ComOrderDetail'--data:4696k

 磁盘io:

set statistics io on
 select * from ComOrderDetail  --0.57
set statistics io off

 

 压缩:

alter table ComOrderDetail rebuild partition=all
	with(data_compression=page)

 压缩后表空间的使用情况:

-- 表空间的使用情况	
sp_spaceused 'ComOrderDetail'  --data:1376k

 

 

对非聚集索引的压缩:

 

 

 

 持久化的计算列:

create table computetable(c1 int,c2 int,c3 as (c1+c2)*50)
	declare @n int
	 set @n=1
	 while @n<50000
	 begin
	 insert computetable values(@n,@n+1)
	 set @n=@n+1
	 end

	 sp_spaceused 'computetable'  --data:1608 KB

	 --cpu 的情况
set statistics time on
 select * from computetable  --0.57
set statistics time off

 

 

 

	create table computetable(c1 int,c2 int,c3 as (c1+c2)*50 persisted)
	declare @n	int
	 set @n=1
	 while @n<50000
	 begin
	 insert computetable values(@n,@n+1)
	 set @n=@n+1
	 end

	 sp_spaceused 'computetable'  --data:1608 KB  1824kb

	 --cpu 的情况
set statistics time on
 select * from computetable  
set statistics time off

 

 

 

 

      

posted @ 2017-01-04 15:56  石shi  阅读(1795)  评论(0编辑  收藏  举报