关于对db_block_gets的理解与实验 实验 一、 自己手动创建的小表 创建一个区大小为 40k SYS@ORCLshow parameter db_block_size NAME TYPE VALUE ------------------------------------ ----------- ------------------------------ db_block_size integer 819
关于对db_block_gets的理解与实验
实验
一、 自己手动创建的小表
创建一个区大小为 40k
SYS@ORCL>show parameter db_block_size
NAME TYPE VALUE
------------------------------------ ----------- ------------------------------
db_block_size integer 8192
SYS@ORCL>create tablespace tyger1 datafile '/u01/app/oracle/oradata/ORCL/tyger1.dbf' size 10m
2 extent management local uniform size 40k;
Tablespace created.
SYS@ORCL>create table test_db1(x int) tablespace tyger1;
Table created.
SYS@ORCL>set autotrace on
SYS@ORCL>insert into test_db1 values(1);
1 row created.
Execution Plan
----------------------------------------------------------
-------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
-------------------------------------------------------------------------
| 0 | INSERT STATEMENT | | 1 | 100 | 1 (0)| 00:00:01 |
-------------------------------------------------------------------------
Statistics
----------------------------------------------------------
1 recursive calls
19 db block gets
1 consistent gets
3 physical reads
9 redo size
675 bytes sent via SQL*Net to client
562 bytes received via SQL*Net from client
4 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
1 rows processed
SYS@ORCL>insert into test_db1 values(2);
1 row created.
Execution Plan
----------------------------------------------------------
-------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
-------------------------------------------------------------------------
| 0 | INSERT STATEMENT | | 1 | 100 | 1 (0)| 00:00:01 |
-------------------------------------------------------------------------
Statistics
----------------------------------------------------------
1 recursive calls
3 db block gets
1 consistent gets
0 physical reads
244 redo size
675 bytes sent via SQL*Net to client
562 bytes received via SQL*Net from client
4 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
1 rows processed
2. 创建一个区 大小为80k
SYS@ORCL>create tablespace tyger2 datafile '/u01/app/oracle/oradata/ORCL/tyger2.dbf' size 10m
2 extent management local uniform size 80k;
Tablespace created.
SYS@ORCL>create table test_db2(x int) tablespace tyger2;
Table created.
SYS@ORCL>insert into test_db2 values(1);
1 row created.
Execution Plan
----------------------------------------------------------
-------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
-------------------------------------------------------------------------
| 0 | INSERT STATEMENT | | 1 | 100 | 1 (0)| 00:00:01 |
-------------------------------------------------------------------------
Statistics
----------------------------------------------------------
1 recursive calls
29 db block gets
1 consistent gets
28 physical reads
13 redo size
675 bytes sent via SQL*Net to client
562 bytes received via SQL*Net from client
4 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
1 rows processed
SYS@ORCL>insert into test_db2 values(2);
1 row created.
Execution Plan
----------------------------------------------------------
-------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
-------------------------------------------------------------------------
| 0 | INSERT STATEMENT | | 1 | 100 | 1 (0)| 00:00:01 |
-------------------------------------------------------------------------
Statistics
----------------------------------------------------------
1 recursive calls
3 db block gets
1 consistent gets
0 physical reads
288 redo size
677 bytes sent via SQL*Net to client
562 bytes received via SQL*Net from client
4 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
1 rows processed
结论：对于新创建的表来说，因为创建的是空表就没有对表里的空间进行分配，当插入第一条数据时，就需要对区上的块进行空间分配和对数据字典的一些操作，就会有比较大的db_block_size。如果再次插入数据的话就基本没有对空间的分配啥的，就会有比较少的db_block_size产生。
所以对于extent指定的区大小来说 同样的空表插入同样的数据 db_block_size 可能不同。
对插入更新、删除的实验：
SYS@ORCL>update test_db1 set x=3 where x=1;
1 row updated.
Execution Plan
----------------------------------------------------------
Plan hash value: 2185639234
-------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
-------------------------------------------------------------------------------
| 0 | UPDATE STATEMENT | | 1 | 13 | 2 (0)| 00:00:01 |
| 1 | UPDATE | TEST_DB1 | | | | |
|* 2 | TABLE ACCESS FULL| TEST_DB1 | 1 | 13 | 2 (0)| 00:00:01 |
-------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - filter("X"=1)
Note
-----
- dynamic sampling used for this statement
Statistics
----------------------------------------------------------
28 recursive calls
1 db block gets
11 consistent gets
0 physical reads
388 redo size
678 bytes sent via SQL*Net to client
565 bytes received via SQL*Net from client
4 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
1 rows processed
SYS@ORCL>delete test_db1 where x=2;
1 row deleted.
Execution Plan
----------------------------------------------------------
Plan hash value: 3135214910
-------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
-------------------------------------------------------------------------------
| 0 | DELETE STATEMENT | | 1 | 13 | 2 (0)| 00:00:01 |
| 1 | DELETE | TEST_DB1 | | | | |
|* 2 | TABLE ACCESS FULL| TEST_DB1 | 1 | 13 | 2 (0)| 00:00:01 |
-------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - filter("X"=2)
Note
-----
- dynamic sampling used for this statement
Statistics
----------------------------------------------------------
5 recursive calls
1 db block gets
9 consistent gets
0 physical reads
288 redo size
678 bytes sent via SQL*Net to client
557 bytes received via SQL*Net from client
4 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
1 rows processed
SYS@ORCL>insert into test_db1 values(&x);
Enter value for x: 1
old 1: insert into test_db1 values(&x)
new 1: insert into test_db1 values(1)
1 row created.
。。
SYS@ORCL>commit;
Commit complete.
SYS@ORCL>select * from test_db1;
X
----------
3
1
2
3
4
5
6
7
8
9
19
10
1
11
12
13
14
15
16
17
18
21 rows selected.
SYS@ORCL>alter system flush buffer_cache;
System altered.
SYS@ORCL>update test_db1 set x=21 where x=18;
1 row updated.
Execution Plan
----------------------------------------------------------
Plan hash value: 2185639234
-------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
-------------------------------------------------------------------------------
| 0 | UPDATE STATEMENT | | 1 | 13 | 2 (0)| 00:00:01 |
| 1 | UPDATE | TEST_DB1 | | | | |
|* 2 | TABLE ACCESS FULL| TEST_DB1 | 1 | 13 | 2 (0)| 00:00:01 |
-------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - filter("X"=18)
Note
-----
- dynamic sampling used for this statement
Statistics
----------------------------------------------------------
5 recursive calls
1 db block gets
9 consistent gets
0 physical reads
412 redo size
678 bytes sent via SQL*Net to client
567 bytes received via SQL*Net from client
4 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
1 rows processed
二、对于比较大的表来说
SYS@ORCL>create table test_db1 as select * from dba_objects;
Table created.
SYS@ORCL>insert into test_db1 values('tyger','tyger','tyger',22,23,'tyger','04-SEP-14','04-SEP-14','tyger','t','t','t','t');
1 row created.
Execution Plan
----------------------------------------------------------
-------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
-------------------------------------------------------------------------
| 0 | INSERT STATEMENT | | 1 | 100 | 1 (0)| 00:00:01 |
-------------------------------------------------------------------------
Statistics
----------------------------------------------------------
1 recursive calls
15 db block gets
1 consistent gets
5 physical reads
1144 redo size
677 bytes sent via SQL*Net to client
6 bytes received via SQL*Net from client
4 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
1 rows processed
SYS@ORCL>alter system flush buffer_cache;
System altered.
SYS@ORCL>update test_db1 set OBJECT_NAME='tom' where owner='tyger';
3 rows updated.
Execution Plan
----------------------------------------------------------
Plan hash value: 2185639234
-------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
-------------------------------------------------------------------------------
| 0 | UPDATE STATEMENT | | 8 | 6 | 154 (2)| 00:00:02 |
| 1 | UPDATE | TEST_DB1 | | | | |
|* 2 | TABLE ACCESS FULL| TEST_DB1 | 8 | 6 | 154 (2)| 00:00:02 |
-------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - filter("OWNER"='tyger')
Note
-----
- dynamic sampling used for this statement
Statistics
----------------------------------------------------------
5 recursive calls
3 db block gets
769 consistent gets
687 physical reads
824 redo size
679 bytes sent via SQL*Net to client
5 bytes received via SQL*Net from client
4 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
3 rows processed
SYS@ORCL>delete test_db1 where owner='tyger';
3 rows deleted.
Execution Plan
----------------------------------------------------------
Plan hash value: 3135214910
-------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
-------------------------------------------------------------------------------
| 0 | DELETE STATEMENT | | 8 | 136 | 154 (2)| 00:00:02 |
| 1 | DELETE | TEST_DB1 | | | | |
|* 2 | TABLE ACCESS FULL| TEST_DB1 | 8 | 136 | 154 (2)| 00:00:02 |
-------------------------------------------------------------------------------
Predicate Information (identified by operation id):
---------------------------------------------------
2 - filter("OWNER"='tyger')
Note
-----
- dynamic sampling used for this statement
Statistics
----------------------------------------------------------
4 recursive calls
3 db block gets
769 consistent gets
0 physical reads
10 redo size
679 bytes sent via SQL*Net to client
567 bytes received via SQL*Net from client
4 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
3 rows processed
结论：对于占用多个段的大表来说，可能对数据修改时 对 数据字典 或者对于区、块的分配都包含在 physical reads中。
感想：
对于生产库来说，这个值一般不会太考虑到底数字是怎么来的，因为数字都比较大，一般只在乎它的大小数量级。
下载本文
