oracle hint用法测试

oracle hint测试

创建测试表

javascript 复制代码
CREATE TABLE CUST
   (CUST_ID NUMBER,
	LAST_NAME VARCHAR2(30),
	FIRST_NAME VARCHAR2(30)
   ) ;

插入1000行数据

javascript 复制代码
DECLARE v_cust_id NUMBER;
v_last_name VARCHAR2(30);
v_first_name VARCHAR(30);
BEGIN FOR i IN 1..1000 LOOP v_cust_id := i;
v_last_name := 'Last' || TO_CHAR(i, 'FM000');
v_first_name := 'First' || TO_CHAR(i, 'FM000');
INSERT into
    cust (cust_id, last_name, first_name)
VALUES
    (v_cust_id, v_last_name, v_first_name);
END LOOP;
COMMIT;
END;

分别创建索引

javascript 复制代码
create index index_custid on cust(cust_id);
create index index_LAST_NAME on cust(LAST_NAME);

进行统计信息收集

javascript 复制代码
EXEC DBMS_STATS.GATHER_DATABASE_STATS  全库统计 
exec dbms_stats.gather_table_stats(ownname = > 'SYS', tabname = > 'CUST'); 单表统计
select LAST_ANALYZED,table_name from dba_tables where table_name = 'CUST'; 查询统计信息

开启执行计划

javascript 复制代码
// An highlighted block
set autotrace on

测试

javascript 复制代码
select * from cust WHERE LAST_NAME ='Last1000' AND  CUST_ID='1000';

结果

bash 复制代码
  CUST_ID LAST_NAME			  FIRST_NAME
---------- ------------------------------ ------------------------------
      1000 Last1000			  First109


Execution Plan
----------------------------------------------------------
Plan hash value: 1939667504

----------------------------------------------------------------------------------------------------
| Id  | Operation			    | Name	   | Rows  | Bytes | Cost (%CPU)| Time	   |
----------------------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT		    |		   |	 1 |	21 |	 2   (0)| 00:00:01 |
|*  1 |  TABLE ACCESS BY INDEX ROWID BATCHED| CUST	   |	 1 |	21 |	 2   (0)| 00:00:01 |
|*  2 |   INDEX RANGE SCAN		    | INDEX_CUSTID |	 1 |	   |	 1   (0)| 00:00:01 |
----------------------------------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

   1 - filter("LAST_NAME"='Last1000')
   2 - access("CUST_ID"=1000)


Statistics
----------------------------------------------------------
	  1  recursive calls
	  0  db block gets
	  5  consistent gets
	  0  physical reads
	  0  redo size
	726  bytes sent via SQL*Net to client
	428  bytes received via SQL*Net from client
	  2  SQL*Net roundtrips to/from client
	  0  sorts (memory)
	  0  sorts (disk)
	  1  rows processed

结果该条sql 走了INDEX_CUSTID 索引

使用hint走index_LAST_NAME 索引

bash 复制代码
select /*+INDEX(cust index_LAST_NAME) */ * from cust WHERE LAST_NAME ='Last1000' AND  CUST_ID='1000';
   CUST_ID LAST_NAME			  FIRST_NAME
---------- ------------------------------ ------------------------------
      1000 Last1000			  First109


Execution Plan
----------------------------------------------------------
Plan hash value: 3778465092

-------------------------------------------------------------------------------------------------------
| Id  | Operation			    | Name	      | Rows  | Bytes | Cost (%CPU)| Time     |
-------------------------------------------------------------------------------------------------------
|   0 | SELECT STATEMENT		    |		      |     1 |    21 |     2	(0)| 00:00:01 |
|*  1 |  TABLE ACCESS BY INDEX ROWID BATCHED| CUST	      |     1 |    21 |     2	(0)| 00:00:01 |
|*  2 |   INDEX RANGE SCAN		    | INDEX_LAST_NAME |     1 |       |     1	(0)| 00:00:01 |
-------------------------------------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

   1 - filter("CUST_ID"=1000)
   2 - access("LAST_NAME"='Last1000')


Statistics
----------------------------------------------------------
	  0  recursive calls
	  0  db block gets
	  5  consistent gets
	  0  physical reads
	  0  redo size
	726  bytes sent via SQL*Net to client
	462  bytes received via SQL*Net from client
	  2  SQL*Net roundtrips to/from client
	  0  sorts (memory)
	  0  sorts (disk)
	  1  rows processed

强制走全表扫测试

bash 复制代码
select /*+ FULL(CUST)*/ * from cust WHERE LAST_NAME ='Last1000' AND  CUST_ID='1000';
  CUST_ID LAST_NAME			  FIRST_NAME
---------- ------------------------------ ------------------------------
      1000 Last1000			  First109


Execution Plan
----------------------------------------------------------
Plan hash value: 260468903

--------------------------------------------------------------------------
| Id  | Operation	  | Name | Rows  | Bytes | Cost (%CPU)| Time	 |
--------------------------------------------------------------------------
|   0 | SELECT STATEMENT  |	 |     1 |    21 |     3   (0)| 00:00:01 |
|*  1 |  TABLE ACCESS FULL| CUST |     1 |    21 |     3   (0)| 00:00:01 |
--------------------------------------------------------------------------

Predicate Information (identified by operation id):
---------------------------------------------------

   1 - filter("LAST_NAME"='Last1000' AND "CUST_ID"=1000)


Statistics
----------------------------------------------------------
	  1  recursive calls
	  0  db block gets
	  6  consistent gets
	  0  physical reads
	  0  redo size
	726  bytes sent via SQL*Net to client
	445  bytes received via SQL*Net from client
	  2  SQL*Net roundtrips to/from client
	  0  sorts (memory)
	  0  sorts (disk)
	  1  rows processed
相关推荐
yoothey22 分钟前
MySQL事务机制解析 - 面试高分知识点
数据库·mysql·面试
Lkstar1 小时前
万字长文Query改写与多路召回实战|从HyDE到RRF融合,召回率提升22%的完整方案
数据库·人工智能·llm
IT新视界1 小时前
星环科技ArgoDB:基于一体化架构构建数据全生命周期安全底座
数据库·科技·安全·架构
峥无1 小时前
MySQL DML 操作(CRUD)总结
数据库·mysql
数据库小学妹2 小时前
SQL Server数据库同步工具怎么选?6款方案对比+信创迁移避坑清单
数据库·经验分享·sqlserver·dba
不剪发的Tony老师2 小时前
国产数据库之GaussDB:固若金汤
数据库·gaussdb
雨辰AI3 小时前
生产级实测:SpringBoot3 + 达梦数据库接口从 200ms 优化至 20ms 完整调优指南
java·数据库·spring boot·后端·政务
凡人叶枫3 小时前
Effective C++ 条款39:明智而审慎地使用 private 继承
java·数据库·c++·嵌入式开发
基德爆肝c语言3 小时前
MySQL表的操作
前端·数据库·mysql
TDengine (老段)4 小时前
TDengine 连接算子 — Inner/Outer/ASOF/Window Join 的实现与使用
大数据·数据库·物联网·哈希算法·时序数据库·tdengine·涛思数据