Re: parallel query

From: Martin Douglas <dafis_at_ppp.test.jccbi.gov>
Date: 1996/10/10
Message-ID: <NEWTNews.845016912.3493.dafis_at_ppp.test.jccbi.gov>#1/1


In article <325ca6a2.2452126_at_n5.gbso.net>, <chuckh_at_dvol.com> writes:
> From: chuckh_at_dvol.com (Chuck Hamilton)
> Newsgroups: comp.databases.oracle
> Subject: parallel query
> Date: Wed, 09 Oct 1996 13:24:13 GMT
> Organization: Do I look organized to you?
> Message-ID: <325ca6a2.2452126_at_n5.gbso.net>
> X-Newsreader: Forte Agent .99f/16.299
> MIME-Version: 1.0
> Content-Type: text/plain; charset=us-ascii
> Content-Transfer-Encoding: 7bit
> NNTP-Posting-Host: 206.20.144.67
> Lines: 12
> Path:

news.ou.edu!news.ecn.uoknor.edu!news.wildstar.net!newsfeed.direct.ca!op.net!new s.mathworks.com!newsfeed.internetmci.com!n5.gbso.net!
>
> I've noticed that even though I don't have an SMP system, if I turn on
> parallel query most of my queries complete about 30% faster.
>
> I've also noticed that in addition to the p_000n background processes
> PQ creates, I also sometimes get additional sessions showing up in
> v$sessions view. What are these sessions and do they count towards my
> licensed users limit? I didn't see anything in the documentation about
> them.
> --
> Chuck Hamilton
> chuckh_at_dvol.com
>
> This message delivered by electronic sled dogs. WOOF!
>
No they are concidered one session to you licencing parameters. These are just the processes spawned by your quiry. Your processor can work alot faster than your disks can push data out. As long as you have more than one storage device for your tablespace your indexes and quiries should work faster. You can set your degree of parallelism equal to the number of devices on your tablespace. That should be a good rule. Just be careful not to globally set all your objects with the same degree of parallelism. It should be set according to your # of devices associated with that object/tablespace. More than one process to a device could actually degrade performance.

Martin Douglas Received on Thu Oct 10 1996 - 00:00:00 CEST

Original text of this message