Interpreting TKPROF results

From: <solbeach_at_cox.net>
Date: Thu, 1 Jul 2004 16:06:36 -0400
Message-Id: <20040701200637.XQDA6456.fed1rmmtao01.cox.net@smtp.west.cox.net>

One portion of our batch processing usually completed in 3- 5 minutes. Now it is taking 50+ minutes & I'm trying to understand where the time is being spent by enabling SQL TRACE. After running the trace file thru TKPROF I see results which have me puzzled. In the past I would have sworn that "cpu" time was always <= "elapsed" time. With more than one SQL statement, I'm seeing "cpu">"elapsed". How do I interpret these number?

SELECT STCODE,ZIPCODE,GEOCODE,CITYNAME,CNTYCODE,DUPL,CURDATE,CURSALERATE, CURUSERATE,CURSPECRATE,PRIORDATE,PRIORSALERATE,PRIORUSERATE,PRIORSPECRATE, CNTYTAXIND,ZIPEXTFIRST,ZIPEXTLAST,ADMNCODE,TAXCODE,EXCPCODE FROM
TAXLOCLTAX WHERE STCODE = :b1 AND ZIPCODE = :b2 AND GEOCODE = :b3

call count cpu elapsed disk query current rows
------- ------ -------- ---------- ---------- ---------- ---------- ----------

Parse        1      0.01       0.00          0          0          0           0
Execute  25620      9.58       2.73          0          0          0           0
Fetch    25620      7.02       2.20          0      76339          0       25099


------- ------  -------- ---------- ---------- ---------- ----------  ----------

total    51241     16.61       4.94          0      76339          0       25099




----------------------------------------------------------------

Please see the official ORACLE-L FAQ: http://www.orafaq.com

To unsubscribe send email to: oracle-l-request_at_freelists.org put 'unsubscribe' in the subject line.
--

Archives are at http://www.freelists.org/archives/oracle-l/ FAQ is at http://www.freelists.org/help/fom-serve/cache/1.html

Received on Thu Jul 01 2004 - 15:03:25 CDT