索引是一个指向表中数据的指针。一个数据库中的索引与一本书的索引目录是非常相似的。
索引可以用来提高数据库查询性能,但是不恰当的使用将导致数据库性能下降。
1.创建表,在表中创建索引
--为表test的testnum列创建一个索引
root@modb:~# su - omm
omm@modb:~$ gsql -r
gsql ((openGauss 3.0.0 build 02c14696) compiled at 2022-04-01 18:12:00 commit 0 last mr )
Non-SSL connection (SSL connection is recommended when requiring high-security)
Type "help" for help.
omm=# drop table if exists test;
NOTICE: table "test" does not exist, skipping
omm=# DROP TABLE
create table test(id serial primary key,testnum serial);
NOTICE: CREATE TABLE will create implicit sequence "test_id_seq" for serial column "test.id"
NOTICE: CREATE TABLE will create implicit sequence "test_testnum_seq" for serial column "test.testnum"
NOTICE: CREATE TABLE / PRIMARY KEY will create implicit index "test_pkey" for table "test"
CREATE TABLE
omm=# create index idx_test_testnum on test(testnum);
CREATE INDEX
--查看索引
omm=# \di
List of relations
Schema | Name | Type | Owner | Table | Storage
--------+------------------+-------+-------+-------+---------
public | idx_test_testnum | index | omm | test |
public | test_pkey | index | omm | test |
(2 rows)
2.通过hint使用索引
--测试准备,创建表customer,并插入数据
omm=# CREATE TABLE customer
omm-# (
omm(# ca_address_sk integer NOT NULL ,
omm(# ca_address_id character(16),
omm(# ca_street_number character(10) ,
omm(# ca_street_name character varying(60) ,
omm(# ca_street_type character(15) ,
omm(# ca_suite_number character(10) ,
omm(# ca_city character varying(60) ,
omm(# ca_county character varying(30) ,
omm(# ca_state character(2) ,
omm(# ca_zip character(10) ,
omm(# ca_country character varying(20) ,
omm(# ca_gmt_offset numeric(5,2) ,
omm(# ca_location_type character(20)
omm(# );
CREATE TABLE
omm=# insert into customer values
(1, 'AAAAAAAABAAAAAAA', '18', 'Jackson', 'Parkway', 'Suite 280', 'Fairfield', 'Maricopa County', 'AZ', '86192' ,'United States', -7.00, 'condo'),
(2, 'AAAAAAAACAAAAAAA', '362', 'Washington 6th', 'RD', 'Suite 80', 'Fairview', 'Taos County', 'NM', '85709', 'United States', -7.00, 'condo'),
(3, 'AAAAAAAADAAAAAAA', '585', 'Dogwood Washington', 'Circle', 'Suite Q', 'Pleasant Valley', 'York County', 'PA', '12477', 'United States', -5.00, 'single family');
INSERT 0 3
--创建索引
omm=# create index customer_idx on customer(ca_address_sk);
CREATE INDEX
--通过hint强制使用索引,查看执行计划
omm=# EXPLAIN SELECT /*+ indexscan(customer customer_idx ) */
* FROM customer WHERE ca_address_sk<100;
QUERY PLAN
-------------------------------------------------------------------------------
[Bypass]
Index Scan using customer_idx on customer (cost=0.00..8.29 rows=2 width=788)
Index Cond: (ca_address_sk < 100)
(3 rows)
3.rename索引
omm=# ALTER INDEX idx_test_testnum RENAME TO idx_test_testnum_new;
ALTER INDEX
4.重建索引
--重建一个单独索引
omm=# ALTER INDEX idx_test_testnum_new REBUILD;
REINDEX
omm=# REINDEX INDEX idx_test_testnum_new;
REINDEX
omm=#
--重建所有索引
omm=# reindex table test;
REINDEX
5.移动索引到其他表空间
--创建表空间myindex_ts:
omm=# CREATE TABLESPACE myindex_ts RELATIVE LOCATION 'tablespace/myindex_ts1';
CREATE TABLESPACE
--将索引idx_test_testnum_new移动到表空间myindex_ts:
omm=# ALTER INDEX idx_test_testnum_new SET TABLESPACE myindex_ts;
ALTER INDEX
--查看索引所在的表空间
omm=# select * from pg_indexes where tablename = 'test';
schemaname | tablename | indexname | tablespace | indexdef
------------+-----------+----------------------+------------+-----------------------------------------------------------------------------------
----
public | test | test_pkey | | CREATE UNIQUE INDEX test_pkey ON test USING btree (id) TABLESPACE pg_default
public | test | idx_test_testnum_new | myindex_ts | CREATE INDEX idx_test_testnum_new ON test USING btree (testnum) TABLESPACE myindex
_ts
(2 rows)
--或
omm=# select * from pg_indexes where indexname = 'idx_test_testnum_new';
schemaname | tablename | indexname | tablespace | indexdef
------------+-----------+----------------------+------------+-----------------------------------------------------------------------------------
----
public | test | idx_test_testnum_new | myindex_ts | CREATE INDEX idx_test_testnum_new ON test USING btree (testnum) TABLESPACE myindex
_ts
(1 row)
6.删除索引
omm=# drop index idx_test_testnum_new;
DROP INDEX
「喜欢这篇文章,您的关注和赞赏是给作者最好的鼓励」
关注作者
【版权声明】本文为墨天轮用户原创内容,转载时必须标注文章的来源(墨天轮),文章链接,文章作者等基本信息,否则作者和墨天轮有权追究责任。如果您发现墨天轮中有涉嫌抄袭或者侵权的内容,欢迎发送邮件至:contact@modb.pro进行举报,并提供相关证据,一经查实,墨天轮将立刻删除相关内容。
评论

2年前

评论
相关阅读
openGauss荣获中国软件行业协会多奖项,技术升级再创行业新高度
openGauss
562次阅读
2025-04-30 14:30:58
MogDB 发布更新,解决 openGauss 数据库在长事务情况下Ustore表膨胀问题
MogDB
309次阅读
2025-04-17 10:41:41
MogDB 发布更新,解决 openGauss 数据库在长事务情况下Ustore表膨胀问题
云和恩墨
202次阅读
2025-04-16 09:52:02
GitCode 成 openGauss 新归宿,国产开源数据库里程碑事件
严少安
176次阅读
2025-04-27 11:37:53
荣誉时刻!openGauss认证证书快递已发,快来看看谁榜上有名!
墨天轮小教习
165次阅读
2025-04-23 17:39:13
单个执行机并行执行MySQL到openGauss数据迁移子任务
Clipnosis
151次阅读
2025-04-30 16:39:58
Postgresql数据库单个Page最多存储多少行数据
maozicb
97次阅读
2025-04-23 16:02:19
openGauss6.0.0适配操作系统自带的软件,不依赖三方库
来杯拿铁
96次阅读
2025-04-18 10:49:53
openGauss新特性 | openGauss-DataVec向量数据库特性介绍
openGauss
68次阅读
2025-04-17 10:41:47
RISC-V 首迎 openGauss 7.0.0-RC1 全量版适配!数据库核心功能完整落地开源架构
openGauss
51次阅读
2025-04-16 10:33:59