Using Oracle Index Hints in SQL statements
Using Oracle Index Hints in SQL statements
Hints are used to give specific information that we know about our data and application, to Oracle. This further improves the performance of our system. There can be instances where the default optimizer may not be efficient for a certain SQL statements. We can specify HINTS with the SQL statements, to improve the efficiency of those SQL statements.
In this article, we shall see how to specify INDEX hints and what the advantages of the same are.
1 How to specify Hints
The Hints are enclosed in comment, /* comment */, in an SQL statement, and can only be specified with SELECT, DELETE and UPDATE keyword.
SELECT /* comment */ ........ ;
All hints should start with a + sign. This tells the SQL Parser that the SQL has a hint specified with it.
SELECT /*+{hint} */ ........ ;
2 Using INDEX Hints
When you specify an INDEX Hint, the optimizer knows that it has to use the INDEX specified in the hint. In this case, the optimizer does not go for a Full Table Scan nor does it use any other index. In addition, it does not calculate the cost of the Index to be used.
If no INDEX hint is specified the optimizer determines the cost of the each index that could be used to access the table and uses the one with the lower cost.
If there are multiple indexes specified with the Hint then the optimizer has to determine the cost of each index to be used with the specified table. Once that is determined, it uses the Index with the lower cost to access the table. In this case, the optimizer does not do a FULL Table Scan. Also note that, the optimizer may choose to use multiple indexes and then merge the result sets to access the table. This method is used if the cost is low.
Syntax:
/*+ INDEX ( table [index [index]...] ) */
Where:
table specifies the name or alias of the table associated with the index to be scanned.
index specifies an index on which an index scan is to be
相关文档:
Oracle笔记
l 关于TRUNC函数
SELECT
RELATED_ID ,
DOC_ID ,
CAT_ID ,
CAT_CODE ,
RELEASE_DATE ,
&n ......
由于以前都是在sqlserver 2005处理,现在客户要求oracle数据库服务器,
最初的代码为:
allRecordSize = (Integer) rs1.getObject(1); //Integer allRecordSize=0;
当执行的时候报:BigDecimal无法转化为Integer类型
为了兼容两者修改后的代码为:
Object o = rs1.getObject(1);
&nbs ......
NULL指的是空值,或者非法值。
NVL (expr1, expr2)->expr1为NULL,返回expr2;不为NULL,返回expr1。注意两者的类型要一致
NVL2 (expr1, expr2, expr3) ->expr1不为NULL,返回expr2;为NULL,返回expr3。expr2和expr3类型不同的话,expr3会转换为expr2的类型
NULLIF (expr1, expr2) ->相等返回NULL,不等返回ex ......
先构造一个表:
create table emp2(
id number(2),
name varchar(10),
currdate date,
action varchar2(1)
)
创建触发器:
create or replace trigger d_i_u_emp2
after insert or update or delete on mysort
begin
if inserting then
insert into emp2 values (12,'dog',sysdate,'i');
elsif deleting then ......