Re: wait events and v$session

From: Daniel Fink <Daniel.Fink_at_Sun.COM>
Date: Tue, 11 May 2004 10:53:52 -0600
Message-id: <40A10520.5040203@sun.com>

Ryan,

Waits are not necessarily caused by concurrency. It would all depend upon the event. For db_file events, waits may be caused by a number of issues. The first that comes to mind is that one block is read from the disk cache, while another has to come from the disk itself, which may also require higher seek times.

The real question is are these waits excessive in either time or number? A q&d calc on sequential_read shows an average of 473 microseconds/wait and 1877 microseconds/wait for scattered (multiblock) reads. How does this compare to what you can expect and/or normally get from your disk subsystem?

Daniel Fink

ryan.gaffuri_at_cox.net wrote:
> I have one batch process running and there are alot of waits. I thought waits were caused by concurrency?
>
> If this is unreadable, please ignore. Hard to format with web email. Basically, I have large waits for DB_FILE_SCATTERED_READ, DB_FILE_SEQUENTIAL_READ, and for RAC Cache Fusion due to the load balancing.
>
> SID EVENT TOTAL_WAITS TOTAL_TIMEOUTS TIME_WAITED AVERAGE_WAIT MAX_WAIT TIME_WAITED_MICRO
> ---------- ------------------------------ ----------- -------------- ----------- ------------ ---------- -----------------
> 22 db file sequential read 491250 0 23237 0 172 232371459
> 22 db file scattered read 159243 0 29892 0 37 298924960

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 Tue May 11 2004 - 11:53:22 CDT

Re: wait events and v$session_event