Oracle 11g新特性:Result Cache-白红宇的个人博客

Oracle 11g新特性:Result Cache

发布日期：2021-07-16 22:42:43 浏览次数：28 分类：技术文章

本文共 10369 字，大约阅读时间需要 34 分钟。

Oracle 11g新特性:Result Cache之一

在Oracle Database 11g中，Oracle引入了一个令人关注的新特性：结果集缓存（Result Cache）。

顾名思义，这个新特性的含义就是将查询的结果集Cache起来，以便随后相同的查询请求可以直接利用，从而避免了再次查询。今天开始来学习一下这个新特性。

进一步的

Result Cache又可以分为：Server Result Cache 和 Client Result Cache。

前者通过服务器端SGA来缓存结果集，后者通过客户端来缓存结果集。

缓存是提高性能的一个常用手段，可以说在Oracle数据库中，Cache无处不在。

对于Client Result Cache:

在使用OCI应用程序时，可以通过客户端内存来缓存查询的结果集，缓存结果可以在所有session间共享，当查询反复执行时，查询结果可以直接从客户段的缓存中获得，从而极大地提高应用效率。

客户端结果集缓存并不使用服务器端的内存，不会对服务器的内存使用造成影响，这一点和Server Result Cache不同。

同Client Result Cache相关的视图主要有：

SQL> select * from dict where table_name like '%CLIENT_RESULT_CACHE%';

TABLE_NAME COMMENTS

------------------------------ ---------------------------------------------

CLIENT_RESULT_CACHE_STATS$ Synonym for CRCSTATS_$

GV$CLIENT_RESULT_CACHE_STATS Synonym for GV_$CLIENT_RESULT_CACHE_STATS

V$CLIENT_RESULT_CACHE_STATS Synonym for V_$CLIENT_RESULT_CACHE_STATS

同Client Result Cache相关的参数有：

SQL> show parameter client_result

NAME TYPE VALUE

------------------------------------ ----------- ------------------------------

client_result_cache_lag big integer 3000

client_result_cache_size big integer 0

如果我们不想启用Client Result Cache的特性，可以设置参数client_result_cache_size为0即可。

对于Server Result Cache:

服务器端结果集缓存使用Shared Pool中的内存来进行结果缓存，这部分内存使用可以通过v$sgastat视图来查询观察：

SQL> select * from v$sgastat

2 where lower(name) like '%result%';

POOL NAME BYTES

------------ -------------------------- ----------

shared pool Result Cache: State Objs 2852

shared pool Result Cache: Memory Mgr 124

shared pool Result Cache: Bloom Fltr 2048

shared pool Result Cache: Cache Mgr 108

进一步的和Result Cache相关的视图有：

SQL> select * from dict where table_name like '%RESULT_CACHE%';

TABLE_NAME COMMENTS

------------------------------ ---------------------------------------------

CLIENT_RESULT_CACHE_STATS$ Synonym for CRCSTATS_$

GV$CLIENT_RESULT_CACHE_STATS Synonym for GV_$CLIENT_RESULT_CACHE_STATS

GV$RESULT_CACHE_DEPENDENCY Synonym for GV_$RESULT_CACHE_DEPENDENCY

GV$RESULT_CACHE_MEMORY Synonym for GV_$RESULT_CACHE_MEMORY

GV$RESULT_CACHE_OBJECTS Synonym for GV_$RESULT_CACHE_OBJECTS

GV$RESULT_CACHE_STATISTICS Synonym for GV_$RESULT_CACHE_STATISTICS

V$CLIENT_RESULT_CACHE_STATS Synonym for V_$CLIENT_RESULT_CACHE_STATS

V$RESULT_CACHE_DEPENDENCY Synonym for V_$RESULT_CACHE_DEPENDENCY

V$RESULT_CACHE_MEMORY Synonym for V_$RESULT_CACHE_MEMORY

V$RESULT_CACHE_OBJECTS Synonym for V_$RESULT_CACHE_OBJECTS

V$RESULT_CACHE_STATISTICS Synonym for V_$RESULT_CACHE_STATISTICS

11 rows selected.

Oracle 11g新特性:Server Result Cache测试

现在我们来看一下Server Result Cache的作用。

首先创建一张测试表：

SQL> connect eygle/eygle

Connected.

SQL> create table eygle as select * from dba_objects;

Table created.

在以前版本中，我们第一次执行该SQL可以看到consistent gets和physical reads大致相同：

SQL> set autotrace on

SQL> select count(*) from eygle;

COUNT(*)

----------

15993

Execution Plan

----------------------------------------------------------

Plan hash value: 3602634261

--------------------------------------------------------------------

| Id | Operation | Name | Rows | Cost (%CPU)| Time |

--------------------------------------------------------------------

| 0 | SELECT STATEMENT | | 1 | 64 (0)| 00:00:01 |

| 1 | SORT AGGREGATE | | 1 | | |

| 2 | TABLE ACCESS FULL| EYGLE | 14489 | 64 (0)| 00:00:01 |

--------------------------------------------------------------------

Note

-----

- dynamic sampling used for this statement

Statistics

----------------------------------------------------------

28 recursive calls

0 db block gets

282 consistent gets

217 physical reads

0 redo size

420 bytes sent via SQL*Net to client

416 bytes received via SQL*Net from client

2 SQL*Net roundtrips to/from client

0 sorts (memory)

0 sorts (disk)

1 rows processed

再次执行同样查询时，由于数据Cache在内存中，physical reads会减少到0.

但是consistent gets仍然不变：

SQL> select count(*) from eygle;

COUNT(*)

----------

15993

Execution Plan

----------------------------------------------------------

Plan hash value: 3602634261

--------------------------------------------------------------------

| Id | Operation | Name | Rows | Cost (%CPU)| Time |

--------------------------------------------------------------------

| 0 | SELECT STATEMENT | | 1 | 64 (0)| 00:00:01 |

| 1 | SORT AGGREGATE | | 1 | | |

| 2 | TABLE ACCESS FULL| EYGLE | 14489 | 64 (0)| 00:00:01 |

--------------------------------------------------------------------

Note

-----

- dynamic sampling used for this statement

Statistics

----------------------------------------------------------

0 recursive calls

0 db block gets

221 consistent gets

0 physical reads

0 redo size

420 bytes sent via SQL*Net to client

416 bytes received via SQL*Net from client

2 SQL*Net roundtrips to/from client

0 sorts (memory)

0 sorts (disk)

1 rows processed

现在我们来看看在Server Result Cache下，Oracle的行为。

首先在result_cache_mode参数设置为MANUAL时：

SQL> show parameter result_cache_mode

NAME TYPE VALUE

------------------------------------ ----------- ------------------------------

result_cache_mode string MANUAL

我们需要在SQL中手工指定Cache，这需要通过加入一个hints来实现，这个hints是result_cache：

SQL> select /*+ result_cache */ count(*) from eygle;

COUNT(*)

----------

15993

Execution Plan

----------------------------------------------------------

Plan hash value: 3602634261

------------------------------------------------------------------------------------------

| Id | Operation | Name | Rows | Cost (%CPU)| Time |

------------------------------------------------------------------------------------------

| 0 | SELECT STATEMENT | | 1 | 64 (0)| 00:00:01 |

| 1 | RESULT CACHE | 76rwwyazv6t6c39f1d8rrqh8rb | | | |

| 2 | SORT AGGREGATE | | 1 | | |

| 3 | TABLE ACCESS FULL| EYGLE | 14489 | 64 (0)| 00:00:01 |

------------------------------------------------------------------------------------------

Result Cache Information (identified by operation id):

------------------------------------------------------

1 - column-count=1; dependencies=(EYGLE.EYGLE);

attributes=(single-row); name="select /*+ result_cache */ count(*) from eygle"

Note

-----

- dynamic sampling used for this statement

Statistics

----------------------------------------------------------

4 recursive calls

0 db block gets

280 consistent gets

0 physical reads

0 redo size

420 bytes sent via SQL*Net to client

416 bytes received via SQL*Net from client

2 SQL*Net roundtrips to/from client

0 sorts (memory)

0 sorts (disk)

1 rows processed

注意到这个执行计划已经和以往的不同，RESULT CACHE以76rwwyazv6t6c39f1d8rrqh8rb名称创建。

那么在接下来的查询中，这个Result Cache就可以被利用：

SQL> select /*+ result_cache */ count(*) from eygle;

COUNT(*)

----------

15993

Execution Plan

----------------------------------------------------------

Plan hash value: 3602634261

------------------------------------------------------------------------------------------

| Id | Operation | Name | Rows | Cost (%CPU)| Time |

------------------------------------------------------------------------------------------

| 0 | SELECT STATEMENT | | 1 | 64 (0)| 00:00:01 |

| 1 | RESULT CACHE | 76rwwyazv6t6c39f1d8rrqh8rb | | | |

| 2 | SORT AGGREGATE | | 1 | | |

| 3 | TABLE ACCESS FULL| EYGLE | 14489 | 64 (0)| 00:00:01 |

------------------------------------------------------------------------------------------

Result Cache Information (identified by operation id):

------------------------------------------------------

1 - column-count=1; dependencies=(EYGLE.EYGLE);

attributes=(single-row); name="select /*+ result_cache */ count(*) from eygle"

Note

-----

- dynamic sampling used for this statement

Statistics

----------------------------------------------------------

0 recursive calls

0 db block gets

0 consistent gets

0 physical reads

0 redo size

420 bytes sent via SQL*Net to client

416 bytes received via SQL*Net from client

2 SQL*Net roundtrips to/from client

0 sorts (memory)

0 sorts (disk)

1 rows processed

在这个利用到Result Cache的查询中，consistent gets减少到0，直接访问结果集，不再需要执行SQL查询。

这就是Result Cache的强大之处。

我们可以通过查询v$result_cache_memory视图来看Cache的使用情况：

SQL> select * from V$RESULT_CACHE_MEMORY

2 where FREE='NO';

ID CHUNK OFFSET FRE OBJECT_ID POSITION

---------- ---------- ---------- --- ---------- ----------

0 0 0 NO 0 0

1 0 1 NO 1 0

通过V$RESULT_CACHE_STATISTICS可以查询Result Cache的统计信息：

SQL> select * from V$RESULT_CACHE_STATISTICS;

ID NAME VALUE

---------- ------------------------------ ----------

1 Block Size (Bytes) 1024

2 Block Count Maximum 992

3 Block Count Current 32

4 Result Size Maximum (Blocks) 49

5 Create Count Success 1

6 Create Count Failure 0

7 Find Count 1

8 Invalidation Count 0

9 Delete Count Invalid 0

10 Delete Count Valid 0

10 rows selected.

V$RESULT_CACHE_OBJECTS记录了Cache的对象：

SQL> SELECT ID,TYPE,NAME,BLOCK_COUNT,ROW_COUNT FROM V$RESULT_CACHE_OBJECTS;

ID TYPE NAME BLOCK_COUNT ROW_COUNT

---------- ---------- ------------------------------ ----------- ----------

0 Dependency EYGLE.EYGLE 1 0

1 Result select /*+ result_cache */ cou 1 1

nt(*) from eygle

一个新的系统包被引入，DBMS_RESULT_CACHE可以用于执行关于Result Cache的管理：

SQL> set serveroutput on

SQL> 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 = 992K bytes (992 blocks)

Maximum Result Size = 49K bytes (49 blocks)

[Memory]

Total Memory = 100836 bytes [0.059% of the Shared Pool]

... Fixed Memory = 5132 bytes [0.003% of the Shared Pool]

... Dynamic Memory = 95704 bytes [0.056% of the Shared Pool]

....... Overhead = 62936 bytes

....... Cache Memory = 32K bytes (32 blocks)

........... Unused Memory = 30 blocks

........... Used Memory = 2 blocks

............... Dependencies = 1 blocks (1 count)

............... Results = 1 blocks

................... SQL = 1 blocks (1 count)

PL/SQL procedure successfully completed.

参考：

查询结果集cache 谷歌

转载地址：https://blog.csdn.net/haiross/article/details/38512971 如侵犯您的版权，请留言回复原文章的地址，我们会给您删除此文章，给您带来不便请您谅解！

上一篇：Oracle 11g全表扫描以Direct Path Read方式执行

下一篇：2014.08.11oracle distinct 原理 /PGA 结果集 Oracle /查询结果集cache

发表评论

关于作者

喝酒易醉，品茶养心，人生如梦，品茶悟道，何以解忧？唯有杜康！

-- 愿君每日到此一游！

Oracle 11g新特性:Result Cache之一

Oracle 11g新特性:Server Result Cache测试

发表评论

最新留言

关于作者

推荐文章