| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
 |
 | |||
| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
|
| |||
Home -> Community -> Mailing Lists -> Oracle-L -> RE: CPU upgrade caused application slow down
The 'SQL*Net message to client' event accounts only for the time
required to push a very small amount of data (typically only 1 byte)
onto a file descriptor. The duration of this event rarely exceeds a few
microseconds. There's no network latency counted in this event's
duration at all.
Cary Millsap
Hotsos Enterprises, Ltd.
http://www.hotsos.com
* Nullius in verba *
Upcoming events:
- Performance Diagnosis 101: 5/7 Dallas, 5/18 New Jersey, 6/22
Pittsburgh
- SQL Optimization 101: 5/3 Boston, 5/24 San Diego, 6/14 Chicago - Hotsos Symposium 2005: March 6-10 Dallas - Visit www.hotsos.com for schedule details...
-----Original Message-----
From: oracle-l-bounce_at_freelists.org
[mailto:oracle-l-bounce_at_freelists.org] On Behalf Of Khedr, Waleed
Sent: Friday, April 30, 2004 11:19 PM
To: oracle-l_at_freelists.org
Subject: RE: CPU upgrade caused application slow down
Imagine you have a session that does a Nested Loop using an index and =
the results are being sent over the network.
If total bytes the needs to be sent is B and the bandwidth is W and the
=
time of execution is T:
Then data pushing rate is B / T
W is constant.
If B / T is < W then transfer time will be B / W=20
Now using faster CPU's T is smaller which causes B / T to be bigger =
value plus the machine is capable of processing more sql requests in the
=
same time.
I'm not sure if this could be reflected as some <sqlnet/data to client>
=
wait event.
-----Original Message-----
From: zhu chao [mailto:chao_ping_at_vip.163.com]
Sent: Friday, April 30, 2004 11:54 PM
To: oracle-l_at_freelists.org
Subject: Re: CPU upgrade caused application slow down
Hi, Waleed:
That is what I do everytime I reorgnized a table. The table that got
reorgnized was an IOT table with Overflow segment. After table =
recreated, I
move overflow segment to make it sorted.
One more thing, the table is not a key table of the applications =
running
on this server, thoug the table is pretty big in size.
I guess the difference in response time is from the network =
layer.But I
cannot prove it. No improvement in response is ok for CPU upgrade, as =
CPU
time is just part of the total response time, but it should not get
longer:(, right?
Thanks
Zhu Chao
> Wondering if you have one major index that gets used most of the time
=
=3D
> using range scan.
> If this is true, then reorganizing the table might have damaged a =
little =3D
> bit the clustering factor.
>
> If this is the case, then loading the table sorted on the columns used
=
=3D
> by that index should help a lot.
>
> Regards,
>
> Waleed
-- Archives are at http://www.freelists.org/archives/oracle-l/ FAQ is at http://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_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 ----------------------------------------------------------------- ---------------------------------------------------------------- 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 Fri Apr 30 2004 - 23:44:36 CDT
 |
 |