从Load profile数据对比来看,应用调整后系统负载有一定的提高,同时每事务逻辑读也有一定的增加(不足10%)。这会是个问题吗?
从第2份数据来看,latch: shared pool 和 latch: library cache有明显的增加。
下面的是性能问题更加严重的时候的AWR数据(采集时长为2小时):
Snap Id Snap Time Sessions Curs/Sess --------- ------------------- -------- --------- Begin Snap: 3744 26-Sep-11 09:00:01 4,504 38.2 End Snap: 3748 26-Sep-11 11:00:26 4,916 52.7 Elapsed: 120.41 (mins) DB Time: 21,868.27 (mins) Cache Sizes ~~~~~~~~~~~ Begin End ---------- ---------- Buffer Cache: 35,840M 35,840M Std Block Size: 8K Shared Pool Size: 6,144M 6,144M Log Buffer: 30,632K Load Profile ~~~~~~~~~~~~ Per Second Per Transaction --------------- --------------- Redo size: 1,790,430.57 5,945.33 Logical reads: 1,162,875.15 3,861.46 Block changes: 154,064.15 511.59 Physical reads: 3,007.80 9.99 Physical writes: 337.21 1.12 User calls: 8,227.28 27.32 Parses: 3,603.24 11.96 Hard parses: 11.25 0.04 Sorts: 7,304.40 24.26 Logons: 11.27 0.04 Executes: 25,994.53 86.32 Transactions: 301.15 % Blocks changed per Read: 13.25 Recursive Call %: 82.87 Rollback per transaction %: 0.21 Rows per Sort: 8.14 Instance Efficiency Percentages (Target 100%) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Buffer Nowait %: 99.99 Redo NoWait %: 100.00 Buffer Hit %: 99.75 In-memory Sort %: 100.00 Library Hit %: 99.94 Soft Parse %: 99.69 Execute to Parse %: 86.14 Latch Hit %: 98.90 Parse CPU to Parse Elapsd %: 14.69 % Non-Parse CPU: 84.51 Shared Pool Statistics Begin End ------ ------ Memory Usage %: 95.48 94.33 % SQL with executions>1: 87.63 74.92 % Memory for SQL w/exec>1: 86.52 74.66 Top 5 Timed Events Avg %Total ~~~~~~~~~~~~~~~~~~ wait Call Event Waits Time (s) (ms) Time Wait Class ------------------------------ ------------ ----------- ------ ------ ---------- latch: shared pool 991,918 274,715 277 20.9 Concurrenc latch: library cache 941,443 232,344 247 17.7 Concurrenc CPU time 212,150 16.2 cursor: pin S ############ 115,049 0 8.8 Other latch: cache buffers lru chain 1,106,570 96,442 87 7.4 Other ------------------------------------------------------------- ^LWait Events DB/Inst: CRM2DB/crm2db1 Snaps: 3744-3748 -> s - second -> cs - centisecond - 100th of a second -> ms - millisecond - 1000th of a second -> us - microsecond - 1000000th of a second -> ordered by wait time desc, waits desc (idle events last) -> %Timeouts: value of 0 indicates value was < .5%. Value of null is truly 0 Avg %Time Total Wait wait Waits Event Waits -outs Time (s) (ms) /txn ---------------------------- -------------- ----- ----------- ------- --------- latch: shared pool 991,918 N/A 274,715 277 0.5 latch: library cache 941,443 N/A 232,344 247 0.4 cursor: pin S 9,095,470,035 N/A 115,049 0 4,180.5 latch: cache buffers lru cha 1,106,570 N/A 96,442 87 0.5 db file sequential read 7,819,285 N/A 49,564 6 3.6 latch: cache buffers chains 511,936 N/A 35,753 70 0.2 latch free 95,257 0 14,609 153 0.0 latch: object queue header o 221,907 N/A 14,378 65 0.1 log file sync 2,156,343 1 11,496 5 1.0 cursor: pin S wait on X 422,320 100 4,603 11 0.2 gc cr grant 2-way 3,526,856 0 3,430 1 1.6 db file scattered read 624,480 N/A 3,083 5 0.3 enq: TX - row lock contentio 5,946 88 2,600 437 0.0 latch: row cache objects 16,983 N/A 2,281 134 0.0 SQL*Net message from dblink 1,284,487 N/A 2,179 2 0.6 gc buffer busy 353,228 0 2,054 6 0.2 gc cr multi block request 2,499,709 0 1,974 1 1.1 gc current block 2-way 1,969,516 0 1,920 1 0.9 SQL*Net more data from clien 1,667,513 N/A 1,914 1 0.8 control file sequential read 2,974,349 N/A 1,735 1 1.4 db file parallel read 230,795 N/A 1,645 7 0.1 log file parallel write 1,581,432 N/A 1,303 1 0.7 Backup: sbtbackup 14 N/A 730 52120 0.0 buffer busy waits 143,869 0 488 3 0.1 latch: library cache lock 5,058 N/A 378 75 0.0 enq: HW - contention 22,039 0 370 17 0.0 gc current grant 2-way 424,243 N/A 357 1 0.2 enq: SQ - contention 5,010 1 354 71 0.0 read by other session 50,153 0 339 7 0.0 direct path read 324,226 N/A 305 1 0.1 db file parallel write 207,793 N/A 274 1 0.1 row cache lock 175,301 0 263 1 0.1
本文来自电脑杂谈,转载请注明本文网址:
http://www.pc-fly.com/a/jisuanjixue/article-30053-17.html
不开雷达不装炮弹电磁辐射对于军舰来说已经最低了还被直接找到