oracle学习----逻辑读

1.物理读

当数据块第一次读取到,就会缓存到buffer cache 中,而第二次读取和修改该数据块时就在内存buffer cache

清空数据缓冲区

SQL> alter session set events 'immediate trace name flush_cache';

会话已更改。

SQL> select * from dept;

执行计划
----------------------------------------------------------
Plan hash value: 3383998547

--------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 4 | 120 | 3 (0)| 00:00:01 |
| 1 | TABLE ACCESS FULL| DEPT | 4 | 120 | 3 (0)| 00:00:01 |
--------------------------------------------------------------------------

Note
-----
- dynamic sampling used for this statement (level=2)

统计信息
----------------------------------------------------------
0 recursive calls
0 db block gets
8 consistent gets
6 physical reads 当第一次读取数据时，会从数据块中读取，发生物理读
0 redo size
803 bytes sent via SQL*Net to client
519 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
4 rows processed

SQL> select * from dept;

执行计划
----------------------------------------------------------
Plan hash value: 3383998547

--------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 4 | 120 | 3 (0)| 00:00:01 |
| 1 | TABLE ACCESS FULL| DEPT | 4 | 120 | 3 (0)| 00:00:01 |
--------------------------------------------------------------------------

Note
-----
- dynamic sampling used for this statement (level=2)

统计信息
----------------------------------------------------------
0 recursive calls
0 db block gets
8 consistent gets
0 physical reads 当第二次读取数据时，是从db bufffer cache中读取，不发生物理读
0 redo size
803 bytes sent via SQL*Net to client
519 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
4 rows processed

2.逻辑读

逻辑读分为当前读和一致读

2.1 当前读

当前读即读取数据块当前的最新数据。任何时候在Buffer Cache中都只有一份当前数据块。当前读通常发生在对数据进行修改、删除操作时。这时，进程会给数据加上行级锁，并且标识数据为“脏”数据。

SQL> select * from dept for update;使用for update添加行锁

执行计划
----------------------------------------------------------
Plan hash value: 1522154571

----------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
----------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 4 | 120 | 3 (0)| 00:00:01 |
| 1 | FOR UPDATE | | | | | |
| 2 | BUFFER SORT | | | | | |
| 3 | TABLE ACCESS FULL| DEPT | 4 | 120 | 3 (0)| 00:00:01 |
----------------------------------------------------------------------------

Note
-----
- dynamic sampling used for this statement (level=2)

统计信息
----------------------------------------------------------
4 recursive calls
4 db block gets 由于添加了行锁，产生了当前读 current read=db block gets
15 consistent gets
0 physical reads
884 redo size
941 bytes sent via SQL*Net to client
519 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
4 rows processed

SQL> commit;    提交操作，释放行锁，再来查看当前读

提交完成。

SQL> select * from dept;

执行计划
----------------------------------------------------------
Plan hash value: 3383998547

--------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 4 | 120 | 3 (0)| 00:00:01 |
| 1 | TABLE ACCESS FULL| DEPT | 4 | 120 | 3 (0)| 00:00:01 |
--------------------------------------------------------------------------

Note
-----
- dynamic sampling used for this statement (level=2)

统计信息
----------------------------------------------------------
0 recursive calls
0 db block gets 当前读为0
8 consistent gets
0 physical reads
0 redo size
803 bytes sent via SQL*Net to client
519 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
4 rows processed

2.2 一致读

Oracle是一个多用户系统。当一个会话开始读取数据还未结束读取之前，可能会有其他会话修改它将要读取的数据。如果会话读取到修改后的数据，就会造成数据的不一致。一致性读就是为了保证数据的一致性。在Buffer Cache中的数据块上都会有最后一次修改数据块时的SCN。如果一个事务需要修改数据块中数据，会先在回滚段中保存一份修改前数据和SCN的数据块，然后再更新Buffer Cache中的数据块的数据及其SCN，并标识其为“脏”数据。当其他进程读取数据块时，会先比较数据块上的SCN和自己的SCN。如果数据块上的SCN小于等于进程本身的SCN，则直接读取数据块上的数据；如果数据块上的SCN大于进程本身的SCN，则会从回滚段中找出修改前的数据块读取数据。通常，普通查询都是一致性读。

SQL session1> select * from dept;

执行计划
----------------------------------------------------------
Plan hash value: 3383998547

--------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 4 | 120 | 3 (0)| 00:00:01 |
| 1 | TABLE ACCESS FULL| DEPT | 4 | 120 | 3 (0)| 00:00:01 |
--------------------------------------------------------------------------

Note
-----
- dynamic sampling used for this statement (level=2)

统计信息
----------------------------------------------------------
0 recursive calls
0 db block gets
8 consistent gets 一致读
0 physical reads
0 redo size
803 bytes sent via SQL*Net to client
519 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
4 rows processed

开启一个新的会话session2

SQL session2> select * from dept;

DEPTNO DNAME LOC
---------- -------------- -------------
10 ACCOUNTING NEW YORK
20 RESEARCH DALLAS
30 SALES CHICAGO
40 OPERATIONS BOSTON

SQL session2> update dept set loc='NEW YORKA' where deptno=10; 在session2中进行更新操作，但是不提价，这是再来查询session1

已更新 1 行。

SQL session1> select * from dept;

执行计划
----------------------------------------------------------
Plan hash value: 3383998547

--------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 4 | 120 | 3 (0)| 00:00:01 |
| 1 | TABLE ACCESS FULL| DEPT | 4 | 120 | 3 (0)| 00:00:01 |
--------------------------------------------------------------------------

Note
-----
- dynamic sampling used for this statement (level=2)

统计信息
----------------------------------------------------------
0 recursive calls
0 db block gets
9 consistent gets 由于session2中的更新操作没有提交，导致查询数据是，需要构建CR块，一致读增长
0 physical reads
0 redo size
803 bytes sent via SQL*Net to client
519 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
4 rows processed

SQL session2> rollback; 在session2中回滚更新的数据，再在session1中查询，观察一致读

回退已完成。

SQL session1> select * from dept;

执行计划
----------------------------------------------------------
Plan hash value: 3383998547

--------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
--------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 4 | 120 | 3 (0)| 00:00:01 |
| 1 | TABLE ACCESS FULL| DEPT | 4 | 120 | 3 (0)| 00:00:01 |
--------------------------------------------------------------------------

Note
-----
- dynamic sampling used for this statement (level=2)

统计信息
----------------------------------------------------------
0 recursive calls
0 db block gets
8 consistent gets一致读回归
0 physical reads
0 redo size
803 bytes sent via SQL*Net to client
519 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
0 sorts (memory)
0 sorts (disk)
4 rows processed