Re: Asynchronous I/O and AIX and jfs = perf ?

I had the same question as Doug and I received this information from Oracle tech support ...

"Unlike some other unix platforms where Asynchronous I/O is only applicable to raw devices on AIX systems it is applicable to datafiles on both Unix File systemsand Raw Devices. AS such Asynchronous I/O *MUST* be "available for an oracle instance to run properly. If you do NOT have async i/o configured, undefined symbols such as "kaio_rdwr, iosuspend, acancel..." etc. will be flagged during relinking the rdbms. Async I/O is configured by running the root pre.sh script found inthe $ORACLE_HOME/orainst directory. If ct wants to take advantage of AIO for Oracle he must set USE_ASYNC_IO=TRUE"

The only concern should be note:73397.1 which describes a possible problem with async i/o and jfs.

-Keith

In article <385FE5F4.5278DD31_at_bigfoot.com>,   Doug Cowles <dcowles_at_bigfoot.com> wrote:
> Not to be cute.. but can you point me to any resources or white
papers on
> this
> very subject? Or can you elaborate more? I'd really like to know,
since
> we
> are using asych io on JFS on one of our systems. Any help would be
> GREATLY appreciated...
>
> - Dc.
>
> epikur_at_my-deja.com wrote:
>
> > In article <385D4887.9482C816_at_nospambigfoot.com>,
> > Doug Cowles <dcowles_at_nospambigfoot.com> wrote:
> > > This is a multi-part message in MIME format.
> > > --------------E01C07E2F296D884F552A6BF
> > > Content-Type: text/plain; charset=us-ascii
> > > Content-Transfer-Encoding: 7bit
> > >
> > > But can we go back to the point the first reponder made? Using
> > > Asynch I/O with JFS makes no difference at all? Is that true?
> > >
> > > - Dc.
> >
> > yes, it is true. aio without raw devices is completely senseless. it
> > only eats memory.
> >
> > best wishes
> >
> > > >
> > > > Emmanuel Cortes wrote:
> > > > >
> > > > > To increase performance of oracle database 7.3.x.x, is it
better
> > to use
> > > > > Asynchronous I/O on AIX 4.3.x whith jfs file systems or not ?
> > > > >
> > > > > I read in a redbook (Database Performance on AIX in DB2 UDB
ans
> > Oracle
> > > > > Environnements "SG24-551-00") it's a good thing.
> > > > >
> > > > > But oracle tell if you don't use raw device is not good idea.
> > > > >
> > > > > Where is the truth ?
> > > > >
> > > > > Thanks
> > > > >
> > > > > Emmanuel Cortes
> > > >
> > > > --
> > > > <<<<<<<<<<<<<<<<<<<< * >>>>>>>>>>>>>>>>>>>>
> > > >
> > > > Dan Jennings
> > > > Speaker to Machines
> > > >
> > > > When the only tool you have is a hammer,
> > > > all problems tend to look like nails.
> > > > - Masured
> > > >
> > > > <<<<<<<<<<<<<<<<<<<< * >>>>>>>>>>>>>>>>>>>>
> > >
> > > --------------E01C07E2F296D884F552A6BF
> > > Content-Type: text/x-vcard; charset=us-ascii; name="vcard.vcf"
> > > Content-Transfer-Encoding: 7bit
> > > Content-Description: Card for Doug Cowles
> > > Content-Disposition: attachment; filename="vcard.vcf"
> > >
> > > begin: vcard
> > > fn: Doug Cowles
> > > n: Cowles;Doug
> > > org: IBM
> > > email;internet: dcowles_at_nospambigfoot.com
> > > title: DBA
> > > note: Please remove nospam from e-mail to reply.
> > > x-mozilla-cpt: ;0
> > > x-mozilla-html: FALSE
> > > version: 2.1
> > > end: vcard
> > >
> > > --------------E01C07E2F296D884F552A6BF--
> > >
> > >
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>
>

Sent via Deja.com http://www.deja.com/
Before you buy. Received on Tue Jan 04 2000 - 10:36:33 CST