RE: Confirmation of trace data
- From: "Cary Millsap" <cary.millsap@xxxxxxxxxx>
- To: <oracle-l@xxxxxxxxxxxxx>
- Date: Mon, 2 Aug 2004 23:04:16 -0500
I think it's probably time spent preempted, waiting either in the CPU =
run
queue or swapped out. If it were practically any OS other than Windows, =
it
would be relatively easy to get corroborating evidence from the OS about
involuntary context switch counts, average CPU run queue lengths, =
swap-out
counts, and such as that.
But on Windows, I'm not sure. I guess the search begins at
www.sysinternals.com....
Cary Millsap
Hotsos Enterprises, Ltd.
http://www.hotsos.com
* Nullius in verba *
Upcoming events:
- Performance Diagnosis 101: 8/10 Boston, 9/14 San Francisco, 10/5 =
Charlotte
- SQL Optimization 101: 7/26 Washington DC, 8/16 Minneapolis, 9/20 =
Hartford
- Hotsos Symposium 2005: March 6-10 Dallas
- Visit www.hotsos.com for schedule details...
-----Original Message-----
From: oracle-l-bounce@xxxxxxxxxxxxx =
[mailto:oracle-l-bounce@xxxxxxxxxxxxx]
On Behalf Of Paul Baumgartel
Sent: Monday, August 02, 2004 4:11 PM
To: oracle-l@xxxxxxxxxxxxx
Subject: Confirmation of trace data
List--
Is there anything I'm missing here? The following SQL is taking 6 to 8
seconds of elapsed time to run. There are *no* wait events in the
trace file (all trace data connected with this SQL appears below).=20
Total CPU for the operations is about .12 seconds, yet as you can see,
the elapsed time for each fetch is 6 to 8 seconds.
This SQL is running during a load test; there are approximately 100
other sessions on the server (Windows 2K, dual-hyper-threaded CPU,
about 2 GB of memory allotted to Oracle) and at times the CPUs are
pegged at 100%. Oracle9iR2.
Given that I see no wait events, can I conclude that the elapsed time
is due to waiting for CPU service? Is the fact that I'm joining to
DBA_ROLE_PRIVS problematic? =20
TIA.
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
PARSING IN CURSOR #6 len=3D142 dep=3D2 uid=3D828 oct=3D3 lid=3D828 =
tim=3D3027624861
hv=3D2231874444 ad=3D'6fd24b08'
SELECT t1.schema_name
from tsc_schemata t1, dba_role_privs t2
where t2.grantee =3D :b1 and t1.application_role =3D
t2.granted_role
END OF STMT
PARSE =
#6:c=3D0,e=3D92,p=3D0,cr=3D0,cu=3D0,mis=3D0,r=3D0,dep=3D2,og=3D4,tim=3D30=
27624847
EXEC =
#6:c=3D0,e=3D201,p=3D0,cr=3D0,cu=3D0,mis=3D0,r=3D0,dep=3D2,og=3D4,tim=3D3=
027677087
FETCH
#6:c=3D31250,e=3D8150816,p=3D0,cr=3D53,cu=3D0,mis=3D0,r=3D1,dep=3D2,og=3D=
4,tim=3D3035838225
EXEC =
#6:c=3D0,e=3D101,p=3D0,cr=3D0,cu=3D0,mis=3D0,r=3D0,dep=3D2,og=3D4,tim=3D3=
036090408
FETCH
#6:c=3D93750,e=3D6570233,p=3D0,cr=3D53,cu=3D0,mis=3D0,r=3D1,dep=3D2,og=3D=
4,tim=3D3042667603
=09
__________________________________
Do you Yahoo!?
Take Yahoo! Mail with you! Get it on your mobile phone.
http://mobile.yahoo.com/maildemo=20
----------------------------------------------------------------
Please see the official ORACLE-L FAQ: http://www.orafaq.com
----------------------------------------------------------------
To unsubscribe send email to: oracle-l-request@xxxxxxxxxxxxx
put 'unsubscribe' in the subject line.
--
Archives are at //www.freelists.org/archives/oracle-l/
FAQ is at //www.freelists.org/help/fom-serve/cache/1.html
-----------------------------------------------------------------
----------------------------------------------------------------
Please see the official ORACLE-L FAQ: http://www.orafaq.com
----------------------------------------------------------------
To unsubscribe send email to: oracle-l-request@xxxxxxxxxxxxx
put 'unsubscribe' in the subject line.
--
Archives are at //www.freelists.org/archives/oracle-l/
FAQ is at //www.freelists.org/help/fom-serve/cache/1.html
-----------------------------------------------------------------
Other related posts:
