BlogJava-我的漫漫程序之旅-文章分类-数据库

使用虚拟索引进行数据库优化

々上善若水々 — Tue, 29 Apr 2008 02:20:00 GMT

在数据库优化中，索引的重要性不言而喻。但是，在性能调整过程中，一个索引是否能被查询用到，在索引创建之前是无法确定的，而创建索引是一个代价比较高的操作，尤其是数据量较大的时候。

虚拟索引（Virtual Index）不是物理存在的，它并不会创建实际的索引段，只是在数据字典中加了一个索引的记录，使得优化器能够意识到一个索引的存在，从而判断是否使用该索引作为访问路径。当然，实际上最终查询的访问路径是不会使用该虚拟索引的。

所以，虚拟索引的用处就是用来判断一个索引对于sql的执行计划的影响，尤其是对整个数据库的影响，从而判断是否需要创建物理索引。

oracle文档中并没有提到虚拟索引的创建语法，实际上就是普通索引语法后面加一个nosegment关键字即可，B*Tree index和bitmap index都可以。

不同版本的虚拟索引的特性可能不一样，本文的例子执行环境为：

NING@ning>select * from v$version;

BANNER
----------------------------------------------------------------
Oracle Database 10g Enterprise Edition Release 10.2.0.3.0 - Prod
PL/SQL Release 10.2.0.3.0 - Production
CORE 10.2.0.3.0 Production
TNS for 32-bit Windows: Version 10.2.0.3.0 - Production
NLSRTL Version 10.2.0.3.0 - Production

1.创建虚拟索引

NING@ning>create table test(id int,name varchar2(30));

Table created.

NING@ning>insert into test select rownum,object_name from all_objects where rownum<1001;

1000 rows created.

NING@ning>commit;

Commit complete.

NING@ning>create unique index ix_test on test(id) nosegment;

Index created.

NING@ning>analyze table test compute statistics;

Table analyzed.

2.使用虚拟索引

NING@ning>explain plan for select * from test where id=1;

Explained.

NING@ning>select * from table(dbms_xplan.display());

PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------
Plan hash value: 1357081020

--------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 17 | 3 (0)| 00:00:01 |
|* 1 | TABLE ACCESS FULL| TEST | 1 | 17 | 3 (0)| 00:00:01 |
--------------------------------------------------------------------------

必须设置隐含参数”_use_nosegment_indexes”=true（默认为false)后，CBO才能使用虚拟索引ix_test

NING@ning>alter session set "_use_nosegment_indexes"=true;

Session altered.

NING@ning>select * from table(dbms_xplan.display());

PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------
Plan hash value: 166686173

----------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
----------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 17 | 2 (0)| 00:00:01 |
| 1 | TABLE ACCESS BY INDEX ROWID| TEST | 1 | 17 | 2 (0)| 00:00:01 |
|* 2 | INDEX UNIQUE SCAN | IX_TEST | 1 | | 1 (0)| 00:00:01 |
----------------------------------------------------------------------------

RBO无法使用虚拟索引

NING@ning>alter session set optimizer_mode=rule;

Session altered.

NING@ning>explain plan for select * from test where id=1;

Explained.

NING@ning>select * from table(dbms_xplan.display());

PLAN_TABLE_OUTPUT
-----------------------------------------------------------
Plan hash value: 1357081020

----------------------------------
| Id | Operation | Name |
----------------------------------
| 0 | SELECT STATEMENT | |
|* 1 | TABLE ACCESS FULL| TEST |
--------------------------------------------------------------

RBO使用hint可以使用虚拟索引

NING@ning>explain plan for select /*+ index(test,ix_test)*/* from test where id=1;

Explained.

NING@ning>select * from table(dbms_xplan.display());

PLAN_TABLE_OUTPUT
--------------------------------------------------------------------------
Plan hash value: 166686173

---------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
---------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 1 | 17 | 2 (0)| 00:00:01 |
| 1 | TABLE ACCESS BY INDEX ROWID| TEST | 1 | 17 | 2 (0)| 00:00:01 |
|* 2 | INDEX UNIQUE SCAN | IX_TEST | 1 | | 1 (0)| 00:00:01 |
----------------------------------------------------------------------------

3.虚拟索引的特性

无法执行alter index

NING@ning>alter index ix_test rebuild;
alter index ix_test rebuild
*
ERROR at line 1:
ORA-08114: can not alter a fake index

NING@ning>alter index ix_test rename to ix_test2;
alter index ix_test rename to ix_test2
*
ERROR at line 1:
ORA-08114: can not alter a fake index

不能创建和虚拟索引同名的实际索引

NING@ning>create index ix_test on test(name);
create index ix_test on test(name)
*
ERROR at line 1:
ORA-00955: name is already used by an existing object

可以创建和虚拟索引包含相同列但不同名的实际索引

NING@ning>create index ix_test2 on test(id);

Index created.

在10g使用回收站特性的时候，虚拟索引必须显式drop，或者在drop table后purge table后，才能创建同名的索引

NING@ning>drop table test;

Table dropped.

NING@ning>create unique index ix_test on test2(id);
create unique index ix_test on test2(id)
*
ERROR at line 1:
ORA-00955: name is already used by an existing object

NING@ning>drop index ix_test;
drop index ix_test
*
ERROR at line 1:
ORA-38301: can not perform DDL/DML over objects in Recycle Bin

NING@ning>purge table test;

Table purged.

NING@ning>create unique index ix_test on test2(id);

Index created.

查找系统中已经存在的虚拟索引：

SELECT index_owner, index_name
FROM dba_ind_columns
WHERE index_name NOT LIKE 'BIN$%'
MINUS
SELECT owner, index_name
FROM dba_indexes;

虚拟索引分析并且有效，但是数据字典里查不到结果，估计是oracle内部临时保存了分析结果

々上善若水々 2008-04-29 10:20 发表评论

存储过程练习

々上善若水々 — Thu, 10 Apr 2008 02:11:00 GMT

insert into person values(1,'zdw','zdw','test1')
insert into person values(2,'test','test','test2')
insert into person values(3,'admin','admin','admin3')
/*在存储过程中使用子查询*/

create procedure person_sub_query
(
@id int
)
as
  select * from person where id < (select count(*) from person)
go

execute person_sub_query 2

/*在存储过程中修改参数值，使用多个查询语句：*/
create procedure person_multi_query
(
@id int
)
as
  select @id = (select count(*) from person where id > @id)
  select @id = @id - 1
  select * from person where id = @id
go

execute person_multi_query 1

/*创建表*/
create table person
(
  id int primary key,
  username varchar(50) not null,
  password varchar(20) not null,
  address varchar(200) not null
)
/*增加一条记录的存储过程*/
create procedure proc_person
(
  @id int ,
  @username varchar(50),
  @password varchar(20),
  @address varchar(200)
)
as
insert into person(id,username,password,address) values(@id,@username,@password,@address)

go
/*传值顺序是你声明变量时的顺序*/
exec proc_person 1,'admin','admin','bj'

select * from person;

/*修改数据的存储过程*/
create procedure proc_person_update
(
@id int,
@username varchar(50),
@password varchar(50)
)
as
  update person set username=@username , password=@password where id=@id
go

execute proc_person_update 1,'test','test'

select * from person

/*删除数据的存储过程*/

create procedure proc_person_del
(
@id int
)
as
  delete from person where id = @id
go

execute proc_person_del 1

select * from person

々上善若水々 2008-04-10 10:11 发表评论

常用数据库取前几条数据 sql写法

々上善若水々 — Mon, 24 Dec 2007 03:20:00 GMT

如在ms sqlserver 用此语句：
select top 2 * from test01 就会只显示前2条记录，

MYSQL

select * from your_table where .... limit 2;
//使用limit就可以了.

Oracle的语句:
select * from (select rownum r ,* from test) tt
where tt.r > 50 and tt.r <= 100; (最好用PLSQL游标来解决)

DB2中：
select * from payment fetch first 5 row only --查前5条记录

々上善若水々 2007-12-24 11:20 发表评论

Mysql5.0存储过程初探

々上善若水々 — Fri, 23 Nov 2007 23:51:00 GMT

一.创建存储过程

1.基本语法：

create procedure sp_name()
begin
.........
end

2.参数传递

二.调用存储过程

1.基本语法：call sp_name()
注意：存储过程名称后面必须加括号，哪怕该存储过程没有参数传递

三.删除存储过程

1.基本语法：
drop procedure sp_name//
2.注意事项
(1)不能在一个存储过程中删除另一个存储过程，只能调用另一个存储过程

四.区块，条件，循环

1.区块定义，常用
begin
......
end;
也可以给区块起别名，如：
lable:begin
...........
end lable;
可以用leave lable;跳出区块，执行区块以后的代码
2.条件语句

if 条件 then
statement
else
statement
end if;

3.循环语句
(1).while循环

[label:] WHILE expression DO

statements

END WHILE [label] ;

(2).loop循环

[label:] LOOP

statements

END LOOP [label];

(3).repeat until循环

[label:] REPEAT

statements

UNTIL expression

END REPEAT [label] ;

々上善若水々 2007-11-24 07:51 发表评论

解决Mysql乱码问题

々上善若水々 — Fri, 23 Nov 2007 14:38:00 GMT

修改mysql安装目录下的my.ini文件. 打开找到这一行:

# The following options will be read by MySQL client applications.
# Note that only client applications shipped by MySQL are guaranteed
# to read this section. If you want your own MySQL client program to
# honor these values, you need to specify it as an option during the
# MySQL client library initialization.
# [mysql] default-character-set=gbk mysql

客户端字符集把字符集改成gbk.
再找到这一行:

# The default character set that will be used when a new schema or table is
# created and no character set is defined default-character-set=utf8

这个是创建数据库或表时的默认字符集,用utf8处理国际化
对于以前已经创建好的非gbk字符的数据库,在mysql的安装下
的data目录下,找到你建的相应数据库,进去后找到db.opt文件
用记事本打开,内容改成下面的即可.
default-character-set=utf8 default-collation=utf8 最好用utf8,
因为gbk在处理国际化的时候依然会乱码. 最后一种方法,
你可能临时需要设定字符编码,而又不想改配置文件.
那在执行创建脚本前,加入set names gbk(或utf8)即可.
比如: set names utf8; create table user(略);