just do it
分类: Oracle
2013-10-18 09:38:03
引言:的range,hash,list三大分区可能我们大家在熟悉不过了,但什么每种分区适用于什么场景可能是很多人的疑惑点,那么在选择不同的分区时应该注意什么,为什么这么选,有哪些需要我们加以注意的地方,让我们一一来探索一下吧:)
A.创建range分区,一般用于日期化处理,range分区可以很好的管理基于日期来分区的数据
创建一个新用户ls
SYS@LEO> create user ls identified by ls;
SYS@LEO> grant dba to ls;
SYS@LEO> conn ls/ls
创建样本数据表
create table liusheng (orderid number(10),name varchar2(10),ls_date date);
insert into liusheng values (1,'ls1',to_date('1981-01-02','yyyy-mm-dd'));
insert into liusheng values (1,'ls2',to_date('1998-01-03','yyyy-mm-dd'));
insert into liusheng values (1,'ls3',to_date('1999-01-04','yyyy-mm-dd'));
insert into liusheng values (1,'ls4',to_date('2000-01-05','yyyy-mm-dd'));
insert into liusheng values (1,'ls5',to_date('2000-01-06','yyyy-mm-dd'));
insert into liusheng values (1,'ls6',to_date('2001-01-07','yyyy-mm-dd'));
insert into liusheng values (1,'ls7',to_date('2001-01-08','yyyy-mm-dd'));
insert into liusheng values (1,'ls8',to_date('2002-01-09','yyyy-mm-dd'));
insert into liusheng values (1,'ls9',to_date('2002-01-10','yyyy-mm-dd'));
insert into liusheng values (1,'ls10',to_date('2011-01-11','yyyy-mm-dd'));
创建range分区表
create table liusheng_part
partition by range (ls_date)
(
partition liusheng_part_1999_1 values less than (to_date('1999-01-01','yyyy-mm-dd')) ,
partition liusheng_part_2000_1 values less than (to_date('2000-01-01','yyyy-mm-dd')) ,
partition liusheng_part_2001_1 values less than (to_date('2001-01-01','yyyy-mm-dd')) ,
partition liusheng_part_2002_1 values less than (to_date('2002-01-01','yyyy-mm-dd')) ,
partition liusheng_part_2003_1 values less than (to_date('2003-01-01','yyyy-mm-dd')) ,
partition other values less than (maxvalue)
)
as select * from liusheng;
查看分区
select * from liusheng_part partition (liusheng_part_1999_1);
select * from liusheng_part partition (liusheng_part_2000_1);
select * from liusheng_part partition (liusheng_part_2001_1);
select * from liusheng_part partition (liusheng_part_2002_1);
select * from liusheng_part partition (liusheng_part_2003_1);
select * from liusheng_part partition (other);
B.创建hash分区,利用hash函数打散某列使数据均匀分布,一般用于均衡I/O,缺点数据不容易管理,哈希分区不能DROP、SPLIT 以及MERGE分区
我们创建了拥有10个分区的哈希分区表“LIUSHENG_HASH”
LS@LEO> create table liusheng_hash partition by hash(object_id) partitions 10 as select * from dba_objects;
hash分区所占用的区个数,看每个分区占用的个数都差不多,说明数据还是比较均匀分布的
缺点:hash列上数值不能有太多的重复值,否则会导致数据分布不均匀
select partition_name,count(*) from user_extents where segment_name='LIUSHENG_HASH' group by partition_name;
PARTITION_NAME COUNT(*)
------------------------------ ----------
SYS_P27 3
SYS_P26 3
SYS_P22 2
SYS_P28 3
SYS_P29 2
SYS_P21 2
SYS_P23 3
SYS_P25 3
SYS_P30 2
SYS_P24 3
select count(*) from liusheng_hash;
COUNT(*)
----------
9860
C.创建list分区,一般用于数据可枚举,有限个值,可以考虑列表分区,例如国家名字,按州来分区
创建list分区表,我们按国家来分别存放在不同的州,每个州是一个分区
create table liusheng_list
(city_id NUMBER(5),
city_name VARCHAR2(30),
city_state VARCHAR2(20),
city_amount NUMBER(10)
)
partition by list (city_name)
(
partition asia VALUES('china','japan'),
partition europe VALUES ('germany','italy'),
partition africa VALUES('libya','brazil'),
partition other VALUES(DEFAULT) --默认分区
);
插入数据
insert into liusheng_list values(1,'china','asia',100);
insert into liusheng_list values(2,'germany','europe',101);
insert into liusheng_list values(3,'libya','africa',102);
insert into liusheng_list values(4,'liusheng_city','other',103);
查看数据
LS@LEO> select * from liusheng_list;
CITY_ID COUNTRY_NAME STATE CITY_AMOUNT
---------- ------------------------------ -------------------- -----------
1 china asia 100
2 germany europe 101
3 libya africa 102
4 liusheng_city other 103
小结:测试了上面的三大分区后,我相信现在不仅仅有感性的认识而且也加深了理性的认识,分区是个好东西,合理的利用可以提高我们管理收益(但不一定会提高查询收益),使用不当也会给我们添加许多麻烦,三思而后行是做DBA的一个好习惯。
补充:分区之优势
分区技术实质可以把数据分摊到不同的物理位置,增加I/O负载,提高检索效率。
可用性:分区表可以跨越表空间,而普通表则不然,好处就是如果表的一个分区损坏,其他分区不会受到影响我们只需要修复损坏的分区即可