亚洲春色欧美激情

<form id=WZeJxaBel><nobr id=WZeJxaBel></nobr></form>

<address id=WZeJxaBel><nobr id=WZeJxaBel><nobr id=WZeJxaBel></nobr></nobr></address>

首頁 > 數據庫 > 詳細

oracle時間模型

時間：2014-04-28 03:20:03 阅读：807 評論：0 收藏：0 [點我收藏+]

標簽：des com class http blog div code img style java javascript

Oracle在9i或者早期的版本，對于性能優化方面，主要采用命中率模型，後面的版本，也保留著命中率模型，
比如在awr报告中，Instance Efficiency Percentages (Target 100%)就有关于buffer cache，library cache等相关的命中率。

命中率在性能優化時主要體現在：

當命中率不高時，可以通過參數的調整，提高命中率，從而可以提高系統的處理能力

不過命中率的弊端也顯而易見：

命中率無法看到系統的cpu和IO處理能力的分配。
當系統存在性能問題時,比如：應用並發設計問題，鎖爭用等問題時，命中率無法有效的反映到。

所以，oracle引入了時間模型的概念：

這裏有幾個概念：

響應時間=服務時間+等待時間
从數據庫的角度等价于：數據庫响应时间=cpu时间+等待时间。

elapse time：指时空上的物理时间
db time：db time反映系统整体的资源（cpu，IO）被oracle使用的百分比情况，以user call为视角
cpu time：指cpu的使用的有效时间

Db time= DB CPU+Foreground NO-Idle wait time + DB CPU ON QUEUE

這裏邊：

db cpu即數據庫任务使用cpu的时间
Foreground NO-Idle wait time：非空闲的等待（不包括backgroud进程）
DB CPU ON QUEUE:这里代表的是ready to run，但没有获得cpu运行时间片的在队列中等待时间

下面对數據庫进行cpu密集型的测试：

mamicode.com,码迷

$ cat testcpu.sh 
#!/bin/sh
. /home/oracle/.bash_profile
sqlplus -S /nolog<<EOF
        conn /as sysdba
        declare
                cnt number;
        p number;
        begin
        cnt:=dbms_random.value(1,1000000);
        p:=dbms_random.value(1,1000000);
        while(1>0) loop
            cnt:=mod(cnt*p,1000000);
                end loop;
        end;
        /
EOF
###################################################
cat thread.sh 
#!/bin/sh
for((i=1;i<=16;i++))
do
        /home/oracle/testcpu.sh &
done

mamicode.com,码迷

分別16,32個線程進行壓力測試：

thread    cpu    loadavg(1min)    db time        cpu time    elapse time        
16    100%    16.08        31,138s        31,069s        34.45 (mins)
32    100%    31.81        54,812s        28,014s        30.16 (mins)

这里cpu被完全利用，cpu time大小为30*60*16=28,800s;(上面实验中取的awr时间在30分钟左右)
而db time会随着线程数的增加有所增加，虽然可能没有获得cpu时间，但是在run queue队列上，也会计入db time。

下面看下具體的awr報告：

mamicode.com,码迷

16個並發進程下：
<pre class="brush: sql; title: ; notranslate" title="">
              Snap Id      Snap Time      Sessions Curs/Sess
            --------- ------------------- -------- ---------
Begin Snap:     14551 18-Apr-12 23:00:25        36       2.1
  End Snap:     14552 18-Apr-12 23:34:51        20       2.1
   Elapsed:               34.45 (mins)
   DB Time:              518.97 (mins)

Time Model Statistics           DB/Inst: xxx  Snaps: 14551-14552
-> Total time in database user-calls (DB Time): 31138.3s
-> Statistics including the word "background" measure background process
   time, and so do not contribute to the DB time statistic
-> Ordered by % or DB time desc, Statistic name

Statistic Name                                       Time (s) % of DB Time
------------------------------------------ ------------------ ------------
sql execute elapsed time                             31,137.4        100.0
DB CPU                                               31,068.7         99.8
PL/SQL execution elapsed time                        31,065.0         99.8
parse time elapsed                                        3.4           .0
hard parse elapsed time                                   3.3           .0
hard parse (sharing criteria) elapsed time                0.2           .0
hard parse (bind mismatch) elapsed time                   0.1           .0
DB time                                              31,138.3          N/A
          -------------------------------------------------------------

mamicode.com,码迷

mamicode.com,码迷

32个并发进程下：
              Snap Id      Snap Time      Sessions Curs/Sess
            --------- ------------------- -------- ---------
Begin Snap:     14552 18-Apr-12 23:34:51        20       2.1
  End Snap:     14554 19-Apr-12 00:05:01        52       2.5
   Elapsed:               30.16 (mins)
   DB Time:              913.54 (mins)
Time Model Statistics           DB/Inst: xxx  Snaps: 14552-14554
-> Total time in database user-calls (DB Time): 54812.4s
-> Statistics including the word "background" measure background process
   time, and so do not contribute to the DB time statistic
-> Ordered by % or DB time desc, Statistic name

Statistic Name                                       Time (s) % of DB Time
------------------------------------------ ------------------ ------------
sql execute elapsed time                             54,811.1        100.0
PL/SQL execution elapsed time                        54,776.6         99.9
DB CPU                                               28,013.8         51.1
parse time elapsed                                        2.8           .0
hard parse elapsed time                                   2.6           .0
hard parse (sharing criteria) elapsed time                2.5           .0
hard parse (bind mismatch) elapsed time                   0.2           .0
DB time                                              54,812.4          N/A

mamicode.com,码迷

了解了awr裏的時間模型：那麽接下來:

　　

1，等待事件
　　oracle記錄埋點的事件所花費的時間，不同的版本等待事件略有差異。
　　查询v$system_event来监控到oracle各种等待时间的时间，就可以诊断數據庫的性能瓶颈了。

2，top sql：
　　oracle在sql的执行监控中记录了八个時間：
　　APPLICATION_WAIT_TIME，CONCURRENCY_WAIT_TIME，CLUSTER_WAIT_TIME，USER_IO_WAIT_TIME，PLSQL_EXEC_TIME，JAVA_EXEC_TIME，CPU_TIME，ELAPSED_TIME。
　　oracle可以根據這些時間來監控sql的執行情況。

oracle時間模型,码迷,mamicode.com

oracle時間模型

標簽：des com class http blog div code img style java javascript

原文地址：http://www.cnblogs.com/xpchild/p/3694972.html

踩

(0)

贊

(0)

舉報

評論一句话評論（0）

分享檔案

更多>

2021年07月29日 (22)
2021年07月28日 (40)
2021年07月27日 (32)
2021年07月26日 (79)
2021年07月23日 (29)
2021年07月22日 (30)
2021年07月21日 (42)
2021年07月20日 (16)
2021年07月19日 (90)
2021年07月16日 (35)

周排行

更多

友情鏈接

蘭亭集智國之畫百度統計站長統計阿裏雲 chrome插件新版天聽網

關于我們 - 聯系我們 - 留言反饋

? 2014 mamicode.com 版權所有聯系我們:gaon5@hotmail.com

迷上了代碼！

"I acquaint the inhabitants of Liège of this, that they may understand what fate threatens them if they should assume a similar attitude. A raiding party of hostiles had passed near the fort, and had killed, with particular atrocity, a family of settlers. The man and his wife had been tortured to death, the baby had had its brains beaten out against the trunk of a tree, a very young child had been hung by the wrist tendons to two meat hooks on the walls of the ranch-house, and left there to die. One big boy had had his eyelids and lips and nose cut off, and had been staked down to the ground with his remains of a face lying over a red-ant hole. Only two had [Pg 196]managed to escape,—a child of ten, who had carried his tiny sister in his arms, twenty miles of ca?ons and hills, to the post. Every engine of the English Court was put in motion to prevent the Electoral Prince from coming. Oxford had an interview with Schutz, in which he repeated that it was his applying for the writ to the Lord Chancellor instead of to the queen that had done all the mischief; that her Majesty, had it not been for this untoward incident, would have invited the Prince to come over and spend the summer in England—forgetting, as Schutz observed, that the minute before he had assured him that the queen was too much afraid of seeing any of that family here. He advised Schutz—who could not be convinced that he had done anything irregular in his application, quoting numerous proofs to show that it was the accustomed mode of applying for writs—to avoid appearing again at Court; but Schutz, not seeming disposed to follow that advice, immediately received a positive order to the same effect from the queen through another channel. Schutz, therefore, lost no time in returning to Hanover to justify himself. At the same time, Lord Strafford was instructed to write from the Hague, blaming the conduct of Schutz in applying for the writ in the manner he did, as disrespectful to the queen; for, though strictly legal for an absent peer to make such application, the etiquette was that he should defer it till he could do it personally. Strafford ridiculed the idea of any movement being afoot in favour of the Pretender, and observed that, as to sending him out of the Duke of Lorraine's territory, it was not practicable, because the French king maintained that he had fulfilled the treaty, Lorraine not being any part of France. On the other hand, there were striking signs that the cause[17] of Hanover was in the ascendant. Men who watched the course of events decided accordingly. Marlborough, who so lately had been making court to the Pretender, now wrote from Antwerp, urging the House of Hanover to send over the prince without delay to England; that the state of the queen's health made prompt action necessary; and that the presence of the prince in London would secure the succession without risk, without expense, and without war, and was the likeliest measure of inducing France to abandon its design of assisting the Pretender. Shorty had been frantically trying to catch the Deacon's eye, and was making all manner of winks and warning gestures without avail, for the Deacon answered frankly: "Sure you do," Dodd said, and his voice began to rise. He went to the bed, walked along its length to the window, as he talked, never facing Albin. "You know how to make me feel just fine, no worries at all, no complications, just a nice, simple life. With nothing at all in it, Albin. Nothing at all." J. O., I mean it. "Good-bye, my Bluebell, farewell to you, "Cudn't spare one hand to fetch over fifteen—that's a valiant idea. Now d?an't go loitering; fetch out your cattle afore they're roast beef, git out the horses and all the stock—and souse them ricks wot ?un't burning yit." "Do you know my voice?" asked one of Wells's companions. HoME 亚洲春色欧美激情 ENTER NUMBET 0010www.coin7.com.cn
www.cxwycy.com.cn
metroprs.org.cn
www.rammwbd3.com.cn
rcjfbgn6.com.cn
rqasmgx.com.cn
www.sjzliao.com.cn
www.sparkoni.org.cn
wwpets.org.cn
zjlieying.com.cn