[20190214]11g Query Result Cache RC Latches补充.txt
[20190214]11g Query Result Cache RC Latches补充.txt
--//上午测试链接:http://blog.itpub.net/267265/viewspace-2632907/
--//发现自己的一个错误,另外写一篇帖子更正.
--//顺便复习result cache的相关内容:链接:https://blog.csdn.net/jolly10/article/details/81382644
查看SQL结果高速缓存字典信息
(G)V$RESULT_CACHE_STATISTICS : 列出各种高速缓存设置和内存使用量统计信息
(G)V$RESULT_CACHE_MEMORY : 列出所有内存块和相应的统计信息
(G)V$RESULT_CACHE_OBJECTS: 列出所有对象（高速缓存结果和依赖性）及其属性
(G)V$RESULT_CACHE_DEPENDENCY: 列出高速缓存结果之间的依赖性详细信息及依赖性
dbms_result_cache包可以监视和管理result cache
例如:
dbms_result_cache.flush：清除result cache。
dbms_result_cache.invalidate(owner,name)：使某对象的result cache无效。
dbms_result_cache.status：显示result cache的状态。
dbms_result_cache.memory_report：显示result cache的内存使用状况。
1.环境:
SCOTT@book> @ ver1
PORT_STRING VERSION BANNER
------------------------------ -------------- --------------------------------------------------------------------------------
x86_64/Linux 2.4.xx 11.2.0.4.0 Oracle Database 11g Enterprise Edition Release 11.2.0.4.0 - 64bit Production
SCOTT@book> create table job_times ( sid number, time_ela number);
Table created.
SCOTT@book> create table hc_t ( n number(*,0), v varchar2(200)) ;
Table created.
SCOTT@book> insert into hc_t select level, dbms_random.string('p', 200) from dual connect by level <= 10000;
10000 rows created.
SCOTT@book> create unique index i_hc_t on hc_t(n);
Index created.
SCOTT@book> commit;
Commit complete.
--//分析表略.
create or replace procedure do_rc(
p_iterations in number,p_max in number
) is
l_rowid rowid;
l_n number;
begin
insert into job_times
values (sys_context('userenv', 'sid'), dbms_utility.get_time)
returning rowid into l_rowid;
for i in 1 .. p_iterations
loop
l_n:=trunc(dbms_random.value(1, p_max));
for cur in (select /*+ result_cache */ * from hc_t where n=l_n)
loop
null;
end loop;
end loop;
update job_times set
time_ela=dbms_utility.get_time-time_ela
where rowid=l_rowid;
end;
/
--//注:我加入参数p_max,限制取值范围.
--//为了重复测试建立脚本.
$ cat aa.sql
delete from job_times;
Commit ;
declare
l_job number;
begin
for i in 1 .. 4
loop
dbms_job.submit(
job => l_job,
what => 'do_rc(100000,&&1);'
);
end loop;
end;
/
commit ;
2.开始测试:
SCOTT@book> show parameter result
NAME TYPE VALUE
------------------------------------ ------------ ----------
client_result_cache_lag big integer 3000
client_result_cache_size big integer 0
result_cache_max_result integer 5
result_cache_max_size big integer 1792K
result_cache_mode string MANUAL
result_cache_remote_expiration integer 0
SCOTT@book> exec dbms_result_cache.flush()
PL/SQL procedure successfully completed.
SCOTT@book> set serverout on
SCOTT@book> exec dbms_result_cache.memory_report
R e s u l t C a c h e M e m o r y R e p o r t
[Parameters]
Block Size = 0 bytes
Maximum Cache Size = 0 bytes (0 blocks)
Maximum Result Size = 0 bytes (0 blocks)
[Memory]
Total Memory = 40568 bytes [0.022% of the Shared Pool]
... Fixed Memory = 40568 bytes [0.022% of the Shared Pool]
... Dynamic Memory = 0 bytes [0.000% of the Shared Pool]
PL/SQL procedure successfully completed.
--//我前面测试忽略的result cache的大小.
SCOTT@book> @ aa.sql 10000
4 rows deleted.
Commit complete.
PL/SQL procedure successfully completed.
Commit complete.
SCOTT@book> select count(*),avg(TIME_ELA),sum(TIME_ELA) from job_times ;
COUNT(*) AVG(TIME_ELA) SUM(TIME_ELA)
---------- ------------- -------------
4 4001.5 16006
--//以上我上午测试的结果.大约每个job需要40秒上下.
SCOTT@book> exec dbms_result_cache.memory_report
R e s u l t C a c h e M e m o r y R e p o r t
[Parameters]
Block Size = 1K bytes
Maximum Cache Size = 1792K bytes (1792 blocks)
Maximum Result Size = 89K bytes (89 blocks)
[Memory]
Total Memory = 2003960 bytes [1.111% of the Shared Pool]
... Fixed Memory = 40568 bytes [0.022% of the Shared Pool]
... Dynamic Memory = 1963392 bytes [1.089% of the Shared Pool]
....... Overhead = 128384 bytes
....... Cache Memory = 1792K bytes (1792 blocks)
........... Unused Memory = 0 blocks
........... Used Memory = 1792 blocks
............... Dependencies = 1 blocks (1 count)
............... Results = 1791 blocks
................... SQL = 1791 blocks (1791 count)
PL/SQL procedure successfully completed.
--//实际上我的环境仅仅能容纳1791个结果.也就是我的配置太小,共享池不够大.result_cache_max_result=5,仅仅使用共享池的5%.
SCOTT@book> @ aa.sql 1791
4 rows deleted.
Commit complete.
PL/SQL procedure successfully completed.
Commit complete.
SCOTT@book> select count(*),avg(TIME_ELA),sum(TIME_ELA) from job_times ;
COUNT(*) AVG(TIME_ELA) SUM(TIME_ELA)
---------- ------------- -------------
4 440.5 1762
--//你可以发现这个就与没有做result cache的结果相近了.
--//我重启数据库.通过result cache :RC latch记数也可以验证这个问题.
SCOTT@book> column name format a30
SCOTT@book> select name, gets, misses, sleeps, wait_time from v$latch where name like 'Result Cache%';
NAME GETS MISSES SLEEPS WAIT_TIME
------------------------------ ---------- ---------- ---------- ----------
Result Cache: RC Latch 0 0 0 0
Result Cache: SO Latch 0 0 0 0
Result Cache: MB Latch 0 0 0 0
SCOTT@book> @ aa.sql 1791
4 rows deleted.
Commit complete.
PL/SQL procedure successfully completed.
Commit complete.
SCOTT@book> select count(*),avg(TIME_ELA),sum(TIME_ELA) from job_times ;
COUNT(*) AVG(TIME_ELA) SUM(TIME_ELA)
---------- ------------- -------------
4 432 1728
SCOTT@book> select name, gets, misses, sleeps, wait_time from v$latch where name like 'Result Cache%';
NAME GETS MISSES SLEEPS WAIT_TIME
------------------------------ ---------- ---------- ---------- ----------
Result Cache: RC Latch 405177 3865 10 132
Result Cache: SO Latch 8 0 0 0
Result Cache: MB Latch 0 0 0 0
SCOTT@book> @ aa.sql 10000
4 rows deleted.
Commit complete.
PL/SQL procedure successfully completed.
Commit complete.
SCOTT@book> select count(*),avg(TIME_ELA),sum(TIME_ELA) from job_times ;
COUNT(*) AVG(TIME_ELA) SUM(TIME_ELA)
---------- ------------- -------------
4 3978.25 15913
SCOTT@book> select name, gets, misses, sleeps, wait_time from v$latch where name like 'Result Cache%';
NAME GETS MISSES SLEEPS WAIT_TIME
------------------------------ ---------- ---------- ---------- ----------
Result Cache: RC Latch 1787843 534395 683654 67269002
Result Cache: SO Latch 16 0 0 0
Result Cache: MB Latch 0 0 0 0
