buffer busy wait这个事件表明多个进程正在尝试访问同样的cache buffer中的buffers。 class为file header block的buffer busy wait的等待事件表明文件头的块出现争用。 环境：
SQL> select * from v$version;
BANNER
--------------------------------------------------------------------------------
Oracle Database 11g Enterprise Edition Release 11.2.0.2.0 - 64bit Production
PL/SQL Release 11.2.0.2.0 - Production
CORE 11.2.0.2.0 Production
TNS for Linux: Version 11.2.0.2.0 - Production
NLSRTL Version 11.2.0.2.0 - Production
SQL> 创建表为nologging，分别是testwrite1, testwrite2, testwrite3, testwrite4。
DROP TABLESPACE "SOEDATA" INCLUDING CONTENTS AND DATAFILES; CREATE BIGFILE TABLESPACE "SOEDATA" DATAFILE
'+DATA/dbs101/datafile/soedata01.dbf' SIZE 10485760
AUTOEXTEND ON NEXT 65536 MAXSIZE 33554431M
LOGGING ONLINE PERMANENT BLOCKSIZE 8192
EXTENT MANAGEMENT LOCAL UNIFORM. SIZE 1048576 DEFAULT NOCOMPRESS SEGMENT SPACE MANAGEMENT AUTO; create table testwrite1 (tid number, tcode varchar2(40), name char(2000), name1 char(2000), name2 char(2000), name3 char(1000)) nologging; create table testwrite1 (tid number, tcode varchar2(40), name char(2000), name1 char(2000), name2 char(2000), name3 char(1000)) nologging;
create table testwrite2 (tid number, tcode varchar2(40), name char(2000), name1 char(2000), name2 char(2000), name3 char(1000)) nologging;
create table testwrite3 (tid number, tcode varchar2(40), name char(2000), name1 char(2000), name2 char(2000), name3 char(1000)) nologging;
create table testwrite4 (tid number, tcode varchar2(40), name char(2000), name1 char(2000), name2 char(2000), name3 char(1000)) nologging; shell脚本iotest_write.sh：运行iotest_write.sql。
#!/bin/bash
n=1
while (( $n <= 4 ))
do
sqlplus " soe/[email protected]" @iotest_write.sql $n &
n=$(( n+1 ))
done 脚本iotest_write.sql：direct load数据到testwrite四张表中，每张表load 1,000,000条记录。
set timing on set echo on alter session set events '10046 trace name context forever, level 12'; declare
l_sql varchar2(4000);
begin
l_sql := 'insert into testwrite&1 ' || 'select rownum, to_char(rownum), to_char(rownum), to_char(rownum), to_char(rownum), to_char(rownum)'
|| 'from (select rownum from dual connect by rownum<=1000000)';
execute immediate l_sql;
commit;
end;
/ select 1 from dual; 运行前先truncate testwrite四张表
truncate table testwrite1 ;
truncate table testwrite2 ;
truncate table testwrite3 ;
truncate table testwrite4 ; 运行shell脚本iotest_write.sh。 AWR report输出结果
Load Profile Per Second Per Transaction Per Exec Per Call
~~~~~~~~~~~~ --------------- --------------- ---------- ----------
DB Time(s): 1.6 215.0 0.12 11.70
DB CPU(s): 0.1 7.0 0.00 0.38
Redo size: 21,315.0 2,834,705.7
Logical reads: 2,051.8 272,875.0
Block changes: 214.9 28,584.8
Physical reads: 0.7 97.5
Physical writes: 1,429.4 190,101.7 Top 5 Timed Foreground Events
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Avg
wait % DB
Event Waits Time(s) (ms) time Wait Class
------------------------------ ------------ ----------- ------ ------ ----------
buffer busy waits 2,966 2,191 739 53.6 Concurrenc
direct path write 64,291 1,760 27 43.1 User I/O
DB CPU 134 3.3
db file sequential read 874 15 17 .4 User I/O
Data file init write 265 6 22 .1 User I/O Avg
%Time Total Wait wait Waits % DB
Event Waits -outs Time (s) (ms) /txn time
-------------------------- ------------ ----- ---------- ------- -------- ------
buffer busy waits 2,966 0 2,191 739 156.1 53.6
direct path write 64,291 0 1,760 27 3,383.7 43.1
db file sequential read 874 0 15 17 46.0 .4
Data file init write 265 0 6 22 13.9 .1
db file scattered read 156 0 4 24 8.2 .1
control file sequential re 219 0 2 9 11.5 .1
log file switch (private s 4 0 0 105 0.2 .0
log file sync 13 0 0 23 0.7 .0
control file parallel writ 18 0 0 11 0.9 .0 SQL ordered by Elapsed Time DB/Inst: dbs101/dbs101 Snaps: 802-803
-> Resources reported for PL/SQL code includes the resources used by all SQL
statements called by the code.
-> % Total DB Time is the Elapsed Time of the SQL statement divided
into the Total Database Time multiplied by 100
-> %Total - Elapsed Time as a percentage of Total DB time
-> %CPU - CPU Time as a percentage of Elapsed Time
-> %IO - User I/O Time as a percentage of Elapsed Time
-> Captured SQL account for 98.8% of Total DB Time (s): 4,084
-> Captured PL/SQL account for 98.4% of Total DB Time (s): 4,084 Elapsed Elapsed Time
Time (s) Executions per Exec (s) %Total %CPU %IO SQL Id
---------------- -------------- ------------- ------ ------ ------ -------------
1,006.2 1 1,006.16 24.6 3.2 44.1 dcmcu7kww1184
Module: SQL*Plus
declare l_sql varchar2(4000); begin l_sql := 'insert into testwr
ite4 ' || 'select rownum, to_char(rownum), to_char(rownum), to_char(rownum), to_
char(rownum), to_char(rownum)' || 'from (select rownum from dual connect by ro
wnum<=1000000)'; execute immediate l_sql; commit; end;
Tablespace
------------------------------
Av Av Av Av Buffer Av Buf
Reads Reads/s Rd(ms) Blks/Rd Writes Writes/s Waits Wt(ms)
-------------- ------- ------- ------- ------------ -------- ---------- -------
SOEDATA
1 0 0.0 1.0 119,437 47 5,539 395.7
SYSAUX
591 0 22.5 2.7 290 0 0 0.0
SYSTEM
519 0 15.7 1.0 46 0 8 0.0 Tablespace Filename
------------------------ ----------------------------------------------------
Av Av Av Av Buffer Av Buf
Reads Reads/s Rd(ms) Blks/Rd Writes Writes/s Waits Wt(ms)
-------------- ------- ------- ------- ------------ -------- ---------- -------
PERFSTAT_DATA +DATA/dbs101/datafile/perfstat_data.296.751306195
1 0 0.0 1.0 1 0 0 0.0
SOEDATA +DATA/dbs101/datafile/soedata.294.751909089
1 0 0.0 1.0 119,437 47 5,539 395.7
SYSAUX +DATA/dbs101/sysaux01.dbf
591 0 22.5 2.7 290 0 0 0.0
SYSTEM +DATA/dbs101/system01.dbf Segments by Physical Writes DB/Inst: dbs101/dbs101 Snaps: 802-803
-> Total Physical Writes: 3,611,933
-> Captured Segments account for 100.0% of Total Tablespace Subobject Obj. Physical
Owner Name Object Name Name Type Writes %Total
---------- ---------- -------------------- ---------- ----- ------------ -------
SOE SOEDATA TESTWRITE1 TABLE 904,274 25.04
SOE SOEDATA TESTWRITE4 TABLE 903,377 25.01
SOE SOEDATA TESTWRITE2 TABLE 903,126 25.00
SOE SOEDATA TESTWRITE3 TABLE 900,950 24.94
SYS SYSAUX WRH$_SQL_PLAN TABLE 42 .00
------------------------------------------------------------- Segments by Physical Write Requests DB/Inst: dbs101/dbs101 Snaps: 802-803
-> Total Physical Write Requestss: 119,915
-> Captured Segments account for 99.7% of Total Tablespace Subobject Obj. Phys Write
Owner Name Object Name Name Type Requests %Total
---------- ---------- -------------------- ---------- ----- ------------ -------
SOE SOEDATA TESTWRITE1 TABLE 29,903 24.94
SOE SOEDATA TESTWRITE4 TABLE 29,873 24.91
SOE SOEDATA TESTWRITE2 TABLE 29,870 24.91
SOE SOEDATA TESTWRITE3 TABLE 29,800 24.85
SYS SYSAUX WRH$_SQL_PLAN TABLE 37 .03
------------------------------------------------------------- Segments by Direct Physical Writes DB/Inst: dbs101/dbs101 Snaps: 802-803
-> Total Direct Physical Writes: 3,603,256
-> Captured Segments account for 100.0% of Total Tablespace Subobject Obj. Direct
Owner Name Object Name Name Type Writes %Total
---------- ---------- -------------------- ---------- ----- ------------ -------
SOE SOEDATA TESTWRITE1 TABLE 902,596 25.05
SOE SOEDATA TESTWRITE4 TABLE 901,701 25.02
SOE SOEDATA TESTWRITE2 TABLE 901,445 25.02
SOE SOEDATA TESTWRITE3 TABLE 899,271 24.96
SYS SYSAUX WRH$_ACTIVE_SESSION_ 814636_796 TABLE 24 .00
------------------------------------------------------------- Buffer Wait Statistics DB/Inst: dbs101/dbs101 Snaps: 802-803
-> ordered by wait time desc, waits desc Class Waits Total Wait Time (s) Avg Time (ms)
------------------ ----------- ------------------- --------------
file header block 5,521 2,186 396
data block 8 0 0
segment header 2 0 0 从arw报表中可以看出，最大的等待事件时buffer busy waits，class为file header block。 ASH Report输出结果：
Top User Events DB/Inst: dbs101/dbs101 (Jul 24 16:13 to 16:33) Avg Active
Event Event Class % Event Sessions
----------------------------------- --------------- ---------- ----------
buffer busy waits Concurrency 39.84 1.68
direct path write User I/O 33.75 1.43
CPU + Wait for CPU CPU 3.20 0.14
------------------------------------------------------------- Top Event P1/P2/P3 Values DB/Inst: dbs101/dbs101 (Jul 24 16:13 to 16:33) Event % Event P1 Value, P2 Value, P3 Value % Activity
------------------------------ ------- ----------------------------- ----------
Parameter 1 Parameter 2 Parameter 3
-------------------------- -------------------------- --------------------------
buffer busy waits 39.84 "5","2","13" 39.84
file# block# class# direct path write 33.75 "5","8431712","32" 0.02
file number first dba block cnt log file parallel write 8.95 "1","4","1" 0.85
files blocks requests Data file init write 8.19 "1","256","2147483647" 7.97
count intr timeout db file parallel write 1.93 "1","0","2147483647" 1.93
requests interrupt timeout Top Sessions DB/Inst: dbs101/dbs101 (Jul 24 16:13 to 16:33)
-> '# Samples Active' shows the number of ASH samples in which the session
was found waiting for that particular event. The percentage shown
in this column is calculated with respect to wall clock time
and not total database activity.
-> 'XIDs' shows the number of distinct transaction IDs sampled in ASH
when the session was waiting for that particular event
-> For sessions running Parallel Queries, this section will NOT aggregate
the PQ slave activity into the session issuing the PQ. Refer to
the 'Top Sessions running PQs' section for such statistics. Sid, Serial# % Activity Event % Event
--------------- ---------- ------------------------------ ----------
User Program # Samples Active XIDs
-------------------- ------------------------------ ------------------ --------
578, 1 19.27 buffer busy waits 9.94
SOE [email protected](TNS V1-V3) 504/1,200 [ 42%] 1 direct path write 8.58
435/1,200 [ 36%] 1 770, 5 19.27 buffer busy waits 9.98
SOE [email protected](TNS V1-V3) 506/1,200 [ 42%] 1 direct path write 8.36
424/1,200 [ 35%] 1 195, 3 19.21 buffer busy waits 9.96
SOE [email protected](TNS V1-V3) 505/1,200 [ 42%] 1 direct path write 8.50
431/1,200 [ 36%] 1 388, 1 19.21 buffer busy waits 9.96
SOE [email protected](TNS V1-V3) 505/1,200 [ 42%] 1 direct path write 8.30
421/1,200 [ 35%] 1 575, 1 9.01 log file parallel write 8.95
SYS [email protected](LGWR) 454/1,200 [ 38%] 0 Top Blocking Sessions DB/Inst: dbs101/dbs101 (Jul 24 16:13 to 16:33)
-> Blocking session activity percentages are calculated with respect to
waits on enqueues, latches and "buffer busy" only
-> '% Activity' represents the load on the database caused by
a particular blocking session
-> '# Samples Active' shows the number of ASH samples in which the
blocking session was found active.
-> 'XIDs' shows the number of distinct transaction IDs sampled in ASH
when the blocking session was found active. Blocking Sid (Inst) % Activity Event Caused % Event
-------------------- ---------- ------------------------------ ----------
User Program # Samples Active XIDs
-------------------- ------------------------------ ------------------ --------
960, 3( 1) 16.06 buffer busy waits 16.06
SYS [email protected](W000) 206/1,200 [ 17%] 0 390, 9( 1) 12.64 buffer busy waits 12.64
SYS [email protected](W004) 165/1,200 [ 14%] 0 580, 5( 1) 5.88 buffer busy waits 5.88
SYS [email protected](W003) 77/1,200 [ 6%] 0 1149, 5( 1) 4.75 buffer busy waits 4.75
SYS [email protected](W001) 63/1,200 [ 5%] 0 从ash报表中可以看出，最大的等待事件也是buffer busy waits，等待的块是file id是5。
块号是2。 QUERY得到buffer busy wait的对象： set linesize 120
col block_type for a20
col objn for a25
col otype for a15
col filen for 9999
col blockn for 9999999
col obj for a20
col tbs for a10
select
bbw.cnt,
bbw.obj,
bbw.otype,
bbw.sql_id,
bbw.block_type,
nvl(tbs.name,to_char(bbw.p1)) TBS,
tbs_defs.assm ASSM
from (
select
count(*) cnt,
nvl(object_name,CURRENT_OBJ#) obj,
o.object_type otype,
ash.SQL_ID sql_id,
nvl(w.class,'usn '||to_char(ceil((ash.p3-18)/2))||' '||
decode(mod(ash.p3,2),
1,'header',
0,'block')) block_type,
--nvl(w.class,to_char(ash.p3)) block_type,
ash.p1 p1
from v$active_session_history ash,
( select rownum class#, class from v$waitstat ) w,
all_objects o
where event='buffer busy waits'
and w.class#(+)=ash.p3
and o.object_id (+)= ash.CURRENT_OBJ#
and ash.session_state='WAITING'
and ash.sample_time > sysdate - &minutes/(60*24)
--and w.class# > 18
group by o.object_name, ash.current_obj#, o.object_type,
ash.sql_id, w.class, ash.p3, ash.p1
) bbw,
(select file_id,
tablespace_name name
from dba_data_files
) tbs,
(select
tablespace_name NAME,
extent_management LOCAL,
allocation_type EXTENTS,
segment_space_management ASSM,
initial_extent
from dba_tablespaces
) tbs_defs
where tbs.file_id(+) = bbw.p1
and tbs.name=tbs_defs.name
Order by bbw.cnt
CNT OBJ OTYPE SQL_ID BLOCK_TYPE TBS ASSM
---------- -------------- --------- ------------- -------------------- --------- ------
560 TESTWRITE2 TABLE 73uqrqxgbu6hu file header block SOEDATA AUTO
560 TESTWRITE4 TABLE g83bdwkq7vvd7 file header block SOEDATA AUTO
561 TESTWRITE3 TABLE 2ktanngh5bazp file header block SOEDATA AUTO
563 TESTWRITE1 TABLE 71c8yxa0z57xr file header block SOEDATA AUTO 从这里看出是class为file header block的buffer busy wait。 再从10046 trace文件中看到全部等待在一个块上。可以猜到这是ASSM的bmp管理块。
WAIT #182927391528: nam='buffer busy waits' ela= 1151204 file#=5 block#=2 class#=13 obj#=1 tim=1311496182929374
WAIT #182927391528: nam='buffer busy waits' ela= 2609 file#=5 block#=2 class#=13 obj#=1 tim=1311496182932216
WAIT #182927391528: nam='buffer busy waits' ela= 103 file#=5 block#=2 class#=13 obj#=1 tim=1311496185277262
WAIT #182927391528: nam='buffer busy waits' ela= 79 file#=5 block#=2 class#=13 obj#=1 tim=1311496185915601
WAIT #182927391528: nam='buffer busy waits' ela= 111 file#=5 block#=2 class#=13 obj#=1 tim=1311496186546257
WAIT #182927391528: nam='buffer busy waits' ela= 90 file#=5 block#=2 class#=13 obj#=1 tim=1311496187681644
WAIT #182927391528: nam='buffer busy waits' ela= 97 file#=5 block#=2 class#=13 obj#=1 tim=1311496188014463
WAIT #182927391528: nam='buffer busy waits' ela= 107 file#=5 block#=2 class#=13 obj#=1 tim=1311496189215838
WAIT #182927391528: nam='buffer busy waits' ela= 155 file#=5 block#=2 class#=13 obj#=1 tim=1311496189853705
WAIT #182927391528: nam='buffer busy waits' ela= 36 file#=5 block#=2 class#=13 obj#=1 tim=1311496190052796
WAIT #182927391528: nam='buffer busy waits' ela= 86 file#=5 block#=2 class#=13 obj#=1 tim=1311496192331624
WAIT #182927391528: nam='buffer busy waits' ela= 700 file#=5 block#=2 class#=13 obj#=1 tim=1311496193045278
WAIT #182927391528: nam='buffer busy waits' ela= 80 file#=5 block#=2 class#=13 obj#=1 tim=1311496194926218
WAIT #182927391528: nam='buffer busy waits' ela= 86 file#=5 block#=2 class#=13 obj#=1 tim=1311496198071027
WAIT #182927391528: nam='buffer busy waits' ela= 82 file#=5 block#=2 class#=13 obj#=1 tim=1311496198936753
WAIT #182927391528: nam='buffer busy waits' ela= 87 file#=5 block#=2 class#=13 obj#=1 tim=1311496200432685
WAIT #182927391528: nam='buffer busy waits' ela= 85 file#=5 block#=2 class#=13 obj#=1 tim=1311496201468397
WAIT #182927391528: nam='buffer busy waits' ela= 110 file#=5 block#=2 class#=13 obj#=1 tim=1311496202320686
WAIT #182927391528: nam='buffer busy waits' ela= 302 file#=5 block#=2 class#=13 obj#=1 tim=1311496208721118
WAIT #182927391528: nam='buffer busy waits' ela= 28 file#=5 block#=2 class#=13 obj#=1 tim=1311496208791422
WAIT #182927391528: nam='buffer busy waits' ela= 132 file#=5 block#=2 class#=13 obj#=1 tim=1311496209009280
WAIT #182927391528: nam='buffer busy waits' ela= 72 file#=5 block#=2 class#=13 obj#=1 tim=1311496209206937 dump文件的输出结果：
alter system dump datafile 5 block min 0 block max 2; Start dump data blocks tsn: 9 file#:5 minblk 2 maxblk 2
Block dump from cache:
Dump of buffer cache at level 4 for tsn=9, rdba=2
BH (0xb8f95938) file#: 5 rdba: 0x00000002 (1024/2) class: 13 ba: 0xb84ce000
set: 19 pool: 3 bsz: 8192 bsi: 0 sflg: 3 pwc: 129,19
dbwrid: 0 obj: -1 objn: 1 tsn: 9 afn: 5 hint: f
hash: [0xdb450b18,0x9bf8b148] lru: [0xb8f95b50,0xb8f958f0]
ckptq: [NULL] fileq: [NULL] objq: [0xd0157700,0xb8f95918] objaq: [0xd01576f0,0xb8f95928]
st: XCURRENT md: NULL tch: 144
flags: block_written_once redo_since_read
LRBA: [0x0.0.0] LSCN: [0x0.0] HSCN: [0xffff.ffffffff] HSUB: [3]
cr pin refcnt: 0 sh pin refcnt: 0
Block dump from disk:
buffer tsn: 9 rdba: 0x00000002 (1024/2)
scn: 0x0000.01ad97fe seq: 0x02 flg: 0x04 tail: 0x97fe1d02
frmt: 0x02 chkval: 0x390b type: 0x1d=KTFB Bitmapped File Space Header
Hex dump of block: st=0, typ_found=1
Dump of memory from 0x0000002A96DE6600 to 0x0000002A96DE8600
2A96DE6600 0000A21D 00000002 01AD97FE 04020000 [................]
2A96DE6610 0000390B 00000400 00000080 00BCC000 [.9..............]
2A96DE6620 00000009 00002000 FFFFFFFD 0000007E [..... ......~...]
2A96DE6630 00BCBFFF 00016EF4 00000A8B 01ABC751 [.....n......Q...]
2A96DE6640 00000000 00000000 00000000 00000000 [................]
2A96DE6650 00B77A00 00000080 00000000 00000000 [.z..............]
2A96DE6660 00000000 00000000 00000000 00000000 [................]
Repeat 504 times
2A96DE85F0 00000000 00000000 00000000 97FE1D02 [................]
File Space Header Block:
Header Control:
RelFno: 1024, Unit: 128, Size: 12369920, Flag: 9
AutoExtend: YES, Increment: 8192, MaxSize: 4294967293
Initial Area: 126, Tail: 12369919, First: 93940, Free: 2699
Deallocation scn: 28034897.0
Header Opcode:
Save: No Pending Op
End dump data blocks tsn: 9 file#: 5 minblk 2 maxblk 2 根据网站的解释
http://sites.google.com/site/embtdbo/wait-event-documentation/oracle-buffer-busy-wait class为"file header block"的buffer busy waits等待事件
Most likely extent allocation problems, look at extent size on tablespace and increase
the extent size to there are few extent allocations and less contention on the File
Header Block. then try increasing the "next extent" size in the tablespace.
This wait can happen when lots of extents are being allocated in the tablespace. 尝试不同的方法解决这个等待：
当tablespace的next默认64k，buffer busy waits用了132s。 Elapsed times include waiting on following events:
Event waited on Times Max. Wait Total Waited
---------------------------------------- Waited ---------- ------------
Disk file operations I/O 1447 0.00 0.49
db file sequential read 963 0.02 3.57
CSS initialization 1 0.00 0.00
CSS operation: query 6 0.00 0.00
CSS operation: action 1 0.00 0.00
direct path write 21987 0.16 199.99
control file sequential read 9620 0.18 19.98
KSV master wait 6195 0.08 9.88
ASM file metadata operation 2136 0.03 0.27
kfk: async disk IO 481 0.01 2.04
Data file init write 481 0.06 13.28
db file single write 481 0.01 2.62
control file parallel write 1443 0.05 10.88
rdbms ipc reply 481 0.04 4.78
enq: CF - contention 2 0.01 0.01
os thread startup 1 0.03 0.03
buffer busy waits 119 2.35 132.67
******************************************************************************** 将tablespace的next从默认64k改到1m，buffer busy waits用了138s。
Elapsed times include waiting on following events:
Event waited on Times Max. Wait Total Waited
---------------------------------------- Waited ---------- ------------
Disk file operations I/O 715 0.00 0.25
db file sequential read 475 0.01 1.29
CSS initialization 1 0.00 0.00
CSS operation: query 6 0.00 0.00
CSS operation: action 1 0.00 0.00
direct path write 22599 0.15 197.94
control file sequential read 4740 0.05 9.12
KSV master wait 3042 0.06 5.04
ASM file metadata operation 1047 0.00 0.11
kfk: async disk IO 237 0.01 1.33
Data file init write 237 0.04 6.36
db file single write 237 0.01 1.25
control file parallel write 711 0.07 5.44
rdbms ipc reply 237 0.07 1.98
enq: TX - contention 1 0.00 0.00
enq: CF - contention 2 0.13 0.17
buffer busy waits 140 2.32 138.60
******************************************************************************** 将tablespace的初始大小改到 10g。buffer busy waits没有了。
DROP TABLESPACE "SOEDATA" INCLUDING CONTENTS AND DATAFILES; CREATE BIGFILE TABLESPACE "SOEDATA" DATAFILE
'+DATA/dbs101/datafile/soedata01.dbf' SIZE 10g
AUTOEXTEND ON NEXT 1048576 MAXSIZE 33554431M
LOGGING ONLINE PERMANENT BLOCKSIZE 8192
EXTENT MANAGEMENT LOCAL UNIFORM. SIZE 1048576 DEFAULT NOCOMPRESS SEGMENT SPACE MANAGEMENT AUTO; Elapsed times include waiting on following events:
Event waited on Times Max. Wait Total Waited
---------------------------------------- Waited ---------- ------------
Disk file operations I/O 2 0.00 0.00
CSS initialization 1 0.00 0.00
CSS operation: query 6 0.00 0.00
CSS operation: action 1 0.00 0.00
direct path write 22554 0.14 196.62
KSV master wait 4 0.00 0.00
ASM file metadata operation 2 0.00 0.00
******************************************************************************** 可以看到关于文件扩展的消耗已经没有了。主要的事件是direct path write。
![处理PCX文件[图]](data:image/gif;base64,R0lGODlhAQABAIAAAP///wAAACwAAAAAAQABAAACAkQBADs=)