[转载]---Oracle等待事件2之构造一个DB File Sequential Read等待事件和构造一个Direct Path Read等待事件...

àì夳堔傛蜴生んèń 2021-11-01 06:42 290阅读 0赞

第一篇： [《Oracle等待事件1之分别用表和索引上数据的访问来产生db file scattered read等待事件》][Oracle_1_db file scattered read]

第二篇： [《Oracle等待事件2之构造一个DB File Sequential Read等待事件和构造一个Direct Path Read等待事件》][Oracle_2_DB File Sequential Read_Direct Path Read]

第三篇： 《暂无》

三、构造一个DB File Sequential Read等待时间，等待事件需要在v$session\_wait和10046 trace文件中显示出来，贴出整个演示过程。

db file sequential read等待事件：是由于数据块顺序读产生的，当数据块(索引块)从磁盘一个一个读到内存中时，在这个过程中oracle会发生"db file sequential read"等待事件。

块顺序读场景：索引块顺序读、数据块顺序读、undo回滚构造一致性读、磁盘I/O瓶颈

一般来讲如果检索数据时走索引范围扫描INDEX RANGE SCAN就会发生数据块顺序读的现象，先读取一个索引块，根据索引键值对应ROWID信息在去读ROWID所在的数据块，接下来继续找下一个索引块，在读对应的数据块，就这样一个一个把数据块读取到内存中，这个过程中就会产生"db file sequential read"等待事件。

下面我们来使用索引块顺序读来产生"db file sequential read"

----创建表jack----
      1 SQL> create table jack as select * from dba_objects;
      2 
      3 Table created.
      4 
      ----object_id字段上创建索引，我们要利用索引块顺序读----
      5 SQL> create index jack_ind on jack(object_id);
      6 
      7 Index created.
      8 
      9 SQL> col table_name for a30;
     10 SQL> col index_name for a30;
     11 SQL> select table_name,index_name from user_indexes where table_name='JACK';
     12 
     13 TABLE_NAME               INDEX_NAME
     14 ------------------------------ ------------------------------
     15 JACK                   JACK_IND
     16 
     ----在这里需要收集表数据分布情况，看看有没有数据倾斜，在看看索引键值重复率，表和索引存放的位置和大小等，有了这些信息方便CBO更加客观评价执行计划----
     17 SQL> execute dbms_stats.gather_table_stats(user,'jack',cascade=>true);
     18 
     19 PL/SQL procedure successfully completed.
     20 
     21 SQL> set linesize 150;
     22 SQL> set autotrace trace exp;
     23 
     ----我们查询100-200之间的数据，走的是索引范围扫描，就是先读索引块，通过索引键值找到ROWID，在读数据块，这样一个一个读到内存中，就是数据块顺序读类型----
     24 SQL> select owner,object_id,object_name,object_type,status from jack where object_id>=100 and object_id<=200;
     25 
     26 Execution Plan
     27 ----------------------------------------------------------
     28 Plan hash value: 2860868395
     29 
     30 ----------------------------------------------------------------------------------------
     31 | Id  | Operation            | Name     | Rows  | Bytes | Cost (%CPU)| Time     |
     32 ----------------------------------------------------------------------------------------
     33 |   0 | SELECT STATEMENT        |           |    99 |  5049 |     5     (0)| 00:00:01 |
     34 |   1 |  TABLE ACCESS BY INDEX ROWID| JACK     |    99 |  5049 |     5     (0)| 00:00:01 |
     35 |*  2 |   INDEX RANGE SCAN        | JACK_IND |    99 |       |     2     (0)| 00:00:01 |
     36 ----------------------------------------------------------------------------------------
     37 
     38 Predicate Information (identified by operation id):
     39 ---------------------------------------------------
     40 
     41    2 - access("OBJECT_ID">=100 AND "OBJECT_ID"<=200)
     42 
     43 SQL> set autotrace off;
     44 
     45 SQL> select distinct sid from v$mystat;
     46 
     47     SID
     48 ----------
     49      1
     50 
     51 SQL> select sid,event,total_waits,time_waited from v$session_event;
     52        SID EVENT                                TOTAL_WAITS TIME_WAITED
     53 ---------- ------------------------------------ ----------- -----------
     54      1 Disk file operations I/O                       8       0
     55      1 direct path sync                               2      27
     56      1 enq: RO - fast object reuse                    1       2
     57      1 enq: KO - fast object checkpoint               2       0
     58      1 log file sync                                  1       2
     59      1 db file sequential read                       2165     160
     60      1 db file scattered read                        538      1
     ----在这里我们先标记一下db file sequential read的等待次数，如果一会次数增加那么说明成功触发了这个等待事件----
     61 
      ----打开输出，并创建一个存储过程p2循环800000次(循环的越多执行的时间越长)，在循环的过程中我们可以在v$session_wait视图上捕捉到
     "db file sequential read" select count(*) into echo from jack where object_id>=100 and object_id<=200与上面的sql执行过程类似只是多了一个汇总，也走的是索引范围扫描。----
     62 SQL> set serveroutput on
     63 SQL> create or replace procedure p2
     64   2    as 
     65   3       echo number;
     66   4  begin
     67   5      for i in 1..800000
     68   6      loop
     69   7        select count(*) into echo from jack where object_id>=100 and object_id<=200;
     70   8      end loop;
     71   9    dbms_output.put_line('successfully');
     72  10  end;
     73  11  /
     74 
     75 Procedure created.
     76 
      ----需要先清空一下buffer_cache里面的----
     77 SQL> alter system flush buffer_cache;
     78 
     79 System altered.
     80 
     81 SQL> alter system flush buffer_cache;
     82 
     83 System altered.
     84 
      ----当看到"successfully"表名这个p2执行完毕----
     85 SQL> execute p2;
     86 successfully
     87 
     88 PL/SQL procedure successfully completed.
      ----下面这个查询是执行完上面的存储过程后的一个查询----
     89 SQL> select sid,event,total_waits,time_waited from v$session_event where sid=1;
          SID           EVENT                        TOTAL_WAITS TIME_WAITED
         ------- ----------------------------------- ----------- ----------- 
           1         db file sequential read           2444           22
     ----在执行上面存储过程的时候，在会话31中查询v$session_wait视图，捕捉到"db file sequential read"，如果执行完毕了就看不到这个等待事件。----
     90 SQL> select distinct sid from v$mystat;
     91 
     92     SID
     93 ----------
     94     31
     95 
     96 SQL> set linesize 200;
     97 SQL> col event for a40;
     98 SQL> col p1text for a10;
     99 SQL> col p2text for a10;
    100 SQL> col p3text for a10;
    101 SQL> col wait_class for a20;
    102 SQL> select sid,event,wait_class,p1,p1text,p2,p2text,p3,p3text from v$session_wait where event like '%sequential%';
    103 
    104        SID EVENT                           WAIT_CLASS     P1      P1TEXT        P2      P2TEXT        P3      P3TEXT
    105 ---------- ------------------------------- ---------- ---------- ---------- ---------- ---------- ---------- ----------
    106      1     db file sequential read          User I/O      7       file#         772      block#        1       blocks

从上面的查询可以看到出现了db file sequential read等待事件

小提示：当你进行多次实验后，会发现要访问的trace文件非常的大，在里面找到自己测试sql语句是一件很麻烦的事，这时我们可以选择先清空它再进行测试。

1 SQL> @/u01/scripts/showtrace
     2 
     3 trace_file_name
     4 ----------------------------------------------------------------------------/u01/app/oracle/diag/rdbms/yft/yft/trace/yft_ora_4758.trc
     5 
     6 [oracle@yft ~]$ ll /u01/app/oracle/diag/rdbms/yft/yft/trace/yft_ora_4758.trc
     7 -rw-r--r-- 1 oracle oinstall 1937 Feb 26 09:00 /u01/app/oracle/diag/rdbms/yft/yft/trace/yft_ora_4758.trc
     8 [oracle@yft ~]$ >/u01/app/oracle/diag/rdbms/yft/yft/trace/yft_ora_4758.trc
     9 [oracle@yft ~]$ ll /u01/app/oracle/diag/rdbms/yft/yft/trace/yft_ora_4758.trc
    10 -rw-r--r-- 1 oracle oinstall 0 Feb 26 09:00 /u01/app/oracle/diag/rdbms/yft/yft/trace/yft_ora_4758.trc

10046trace文件显示等待事件：

----清空重新抽取数据----
      1 SQL> alter system flush buffer_cache;
      2 
      3 System altered.
      4 
      ----启动10046事件----
      5 SQL> alter session set events '10046 trace name context forever,level 12';
      6 
      7 Session altered.
      8 
      ----索引范围扫描----
      9 SQL> select count(*) from jack where object_id>=100 and object_id<=200;
     10 
     11   COUNT(*)
     12 ----------
     13     94
     14 
      ----再清空----
     15 SQL> alter system flush buffer_cache;
     16 
     17 System altered.
     18 
     19 SQL> select count(*) from jack where object_id>=100 and object_id<=200;
     20 
     21   COUNT(*)
     22 ----------
     23     94
     24 
     25 SQL> alter system flush buffer_cache;
     26 
     27 System altered.
     28 
      ----插入数据也会产生顺序读----
     29 SQL> insert into jack select * from jack;
     30 
     31 72530 rows created.
     32 
     33 SQL> insert into jack select * from jack;
     34 
     35 145060 rows created.
     36 
      ----关闭10046事件----
     37 SQL> alter session set events '10046 trace name context off';
     38 
     39 Session altered.
     40 
      ----trace文件信息----
     41 select count(*) from jack where object_id>=100 and object_id<=200
     42 END OF STMT
     43 PARSE #13:c=7999,e=24432,p=0,cr=0,cu=0,mis=1,r=0,dep=0,og=1,plh=538454857,tim=1361840628594049
     44 EXEC #13:c=0,e=252,p=0,cr=0,cu=0,mis=0,r=0,dep=0,og=1,plh=538454857,tim=1361840628594604
     45 WAIT #13: nam='SQL*Net message to client' ela= 6 driver id=1650815232 #bytes=1 p3=0 obj#=0 tim=1361840628594748
     46 WAIT #13: nam='db file sequential read' ela= 165 file#=7 block#=771 blocks=1 obj#=74838 tim=1361840628595339
     47 WAIT #13: nam='db file sequential read' ela= 153 file#=7 block#=772 blocks=1 obj#=74838 tim=1361840628595853
     48 FETCH #13:c=1000,e=1219,p=2,cr=2,cu=0,mis=0,r=1,dep=0,og=1,plh=538454857,tim=1361840628596104
     49 STAT #13 id=1 cnt=1 pid=0 pos=1 obj=0 op='SORT AGGREGATE (cr=2 pr=2 pw=0 time=0 us)'
     50 STAT #13 id=2 cnt=94 pid=1 pos=1 obj=74838 op='INDEX RANGE SCAN JACK_IND (cr=2 pr=2 pw=0 time=279 us cost=2 size=495 card=99)'
     51 WAIT #13: nam='SQL*Net message from client' ela= 1336 driver id=1650815232 #bytes=1 p3=0 obj#=74838 tim=1361840628597981
     52 FETCH #13:c=0,e=6,p=0,cr=0,cu=0,mis=0,r=0,dep=0,og=0,plh=538454857,tim=1361840628598117
     53 WAIT #13: nam='SQL*Net message to client' ela= 5 driver id=1650815232 #bytes=1 p3=0 obj#=74838 tim=1361840628598183
     54 WAIT #13: nam='SQL*Net message from client' ela= 1916 driver id=1650815232 #bytes=1 p3=0 obj#=74838 tim=1361840628600139
     ----在上面的信息中因为读取的数据量较少，发生的"db file sequential read"也不是很多
     55  
     56 =====================
     57 PARSING IN CURSOR #17 len=65 dep=0 uid=91 oct=3 lid=91 tim=1361840638457476 hv=2567383207 ad='33614884' sqlid='1n5tj6qchfa57'
     58 select count(*) from jack where object_id>=100 and object_id<=200
     59 END OF STMT
     60 PARSE #17:c=1000,e=232,p=0,cr=0,cu=0,mis=0,r=0,dep=0,og=1,plh=538454857,tim=1361840638457471
     61 EXEC #17:c=0,e=65,p=0,cr=0,cu=0,mis=0,r=0,dep=0,og=1,plh=538454857,tim=1361840638457685
     62 WAIT #17: nam='SQL*Net message to client' ela= 6 driver id=1650815232 #bytes=1 p3=0 obj#=0 tim=1361840638457784
     63 WAIT #17: nam='db file sequential read' ela= 228 file#=7 block#=771 blocks=1 obj#=74838 tim=1361840638458277
     64 WAIT #17: nam='db file sequential read' ela= 59 file#=7 block#=772 blocks=1 obj#=74838 tim=1361840638458477
     65 FETCH #17:c=1000,e=603,p=2,cr=2,cu=0,mis=0,r=1,dep=0,og=1,plh=538454857,tim=1361840638458593
     66 STAT #17 id=1 cnt=1 pid=0 pos=1 obj=0 op='SORT AGGREGATE (cr=2 pr=2 pw=0 time=0 us)'
     67 STAT #17 id=2 cnt=94 pid=1 pos=1 obj=74838 op='INDEX RANGE SCAN JACK_IND (cr=2 pr=2 pw=0 time=186 us cost=2 size=495 card=99)'
     68 WAIT #17: nam='SQL*Net message from client' ela= 263 driver id=1650815232 #bytes=1 p3=0 obj#=74838 tim=1361840638459154
     69 FETCH #17:c=0,e=6,p=0,cr=0,cu=0,mis=0,r=0,dep=0,og=0,plh=538454857,tim=1361840638459234
     70 WAIT #17: nam='SQL*Net message to client' ela= 3 driver id=1650815232 #bytes=1 p3=0 obj#=74838 tim=1361840638459291
     71 WAIT #17: nam='SQL*Net message from client' ela= 1396 driver id=1650815232 #bytes=1 p3=0 obj#=74838 tim=1361840638461079
      ----两次查询出现的等待时间不多
     72 
      ----下面内容是两次插入的操作----
     73 =====================
     74 PARSING IN CURSOR #16 len=35 dep=0 uid=91 oct=2 lid=91 tim=1361840656334367 hv=1953898233 ad='336157f4' sqlid='3g3n5r9u7c7rt'
     75 insert into jack select * from jack
     76 END OF STMT
     77 PARSE #16:c=38994,e=38870,p=5,cr=5,cu=0,mis=1,r=0,dep=0,og=1,plh=949574992,tim=1361840656334362
     78 WAIT #16: nam='db file sequential read' ela= 788 file#=7 block#=130 blocks=1 obj#=74837 tim=1361840656335719
     79 WAIT #16: nam='db file scattered read' ela= 145 file#=7 block#=131 blocks=2 obj#=74837 tim=1361840656336308
     80 WAIT #16: nam='db file sequential read' ela= 180 file#=7 block#=129 blocks=1 obj#=74837 tim=1361840656337231
     81 WAIT #16: nam='db file sequential read' ela= 117 file#=7 block#=640 blocks=1 obj#=74837 tim=1361840656337435
     82 WAIT #16: nam='db file sequential read' ela= 29 file#=3 block#=224 blocks=1 obj#=0 tim=1361840656338534
     83 WAIT #16: nam='db file sequential read' ela= 16 file#=3 block#=292 blocks=1 obj#=0 tim=1361840656338654
     84 WAIT #16: nam='db file sequential read' ela= 51 file#=7 block#=771 blocks=1 obj#=74838 tim=1361840656339324
     85 WAIT #16: nam='db file sequential read' ela= 47 file#=7 block#=772 blocks=1 obj#=74838 tim=1361840656339437
     86 WAIT #16: nam='db file sequential read' ela= 97 file#=3 block#=208 blocks=1 obj#=0 tim=1361840656340117
     87 WAIT #16: nam='db file sequential read' ela= 32 file#=3 
      ----插入数据操作中，95%会发生大量的"db file sequential read"等待事件，偶尔也能看到"db file scattered read"
      ----因为什么呢？我们想一想插入流程是不是：第一、先从数据文件中把数据块读到内存里面，这时会发生大量的"db file sequential read"等待事件，
          第二、再把内存中数据插入到数据文件里。
     88 
      ----第二次插入发生"db file sequential read"的次数就少了，是因为第一次已经把大部分数据加载到内存里了，第二次的时候就可以直接从内存里往磁盘上插入数据了，省略了再从磁盘读这一步。----
     89 =====================
     90 PARSING IN CURSOR #1 len=35 dep=0 uid=91 oct=2 lid=91 tim=1361840662586829 hv=1953898233 ad='336157f4' sqlid='3g3n5r9u7c7rt'
     91 insert into jack select * from jack
     92 END OF STMT
     93 PARSE #1:c=0,e=93,p=0,cr=0,cu=0,mis=0,r=0,dep=0,og=1,plh=949574992,tim=1361840662586825
     94 WAIT #1: nam='db file scattered read' ela= 65 file#=7 block#=131 blocks=2 obj#=74837 tim=1361840662587458
     95 WAIT #1: nam='db file sequential read' ela= 23 file#=7 block#=772 blocks=1 obj#=74838 tim=1361840662588206
     96 WAIT #1: nam='db file scattered read' ela= 23 file#=7 block#=133 blocks=2 obj#=74837 tim=1361840662589060
     97 WAIT #1: nam='db file sequential read' ela= 36 file#=6 block#=687 blocks=1 obj#=74838 tim=1361840662589813
     98 WAIT #1: nam='db file sequential read' ela= 24 file#=7 block#=773 blocks=1 obj#=74838 tim=1361840662591319
     99 WAIT #1: nam='db file sequential read' ela= 20 file#=7 block#=135 blocks=1 obj#=74837 tim=1361840662591858
    100 WAIT #1: nam='db file scattered read' ela= 28 file#=7 block#=136 blocks=2 obj#=74837 tim=1361840662592819
    101 WAIT #1: nam='db file scattered read' ela= 26 file#=7 block#=138 blocks=2 obj#=74837 tim=1361840662594614
    102 WAIT #1: nam='db file sequential read' ela= 40 file#=3 block#=240 blocks=1 obj#=0 tim=1361840662595599
    103 WAIT #1: nam='db file sequential read' ela= 121 file#=6 block#=689 blocks=1 obj#=74838 tim=1361840662595998
    104 WAIT #1: nam='db file scattered read' ela= 26 file#=7 block#=140 blocks=2 obj#=74837 tim=1361840662596640
    105 WAIT #1: nam='db file sequential read' ela= 64 file#=3 block#=192 blocks=1 obj#=0 tim=1361840662597439
    106 WAIT #1: nam='db file sequential read' ela= 25 file#=7 block#=774 blocks=1 obj#=74838 tim=1361840662597764
    107 WAIT #1: nam='db file scattered read' ela= 35 file#=7 block#=142 blocks=2 obj#=74837 tim=1361840662599490
    108 WAIT #1: nam='db file scattered read' ela= 30 file#=7 block#=145 blocks=2 obj#=74837 tim=1361840662601746
    109 WAIT #1: nam='db file sequential read' ela= 25 file#=7 block#=769 blocks=1 obj#=74838 tim=1361840662603010

转载于:https://www.cnblogs.com/Richardzhu/articles/2933149.html

[Oracle_1_db file scattered read]: http://www.cnblogs.com/Richardzhu/articles/2913950.html
[Oracle_2_DB File Sequential Read_Direct Path Read]: http://www.cnblogs.com/Richardzhu/articles/2933149.html