Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Mailing Lists -> Oracle-L -> RE: What can cause boost in LIO/PIO?

RE: What can cause boost in LIO/PIO?

From: Jesse, Rich <Rich.Jesse_at_qtiworld.com>
Date: Fri, 20 Jun 2003 11:46:08 -0700
Message-ID: <F001.005B696A.20030620112535@fatcity.com>


I just got a call from the S/A that the copy completed. After looking back at Spotlight I asked "At 1:00, right?". Sure enough. So at least that part's solved. As to why/how it started the same time as the DBMS_JOB? I'm willing to let that be a mystery pondered on over beers for years and years.

Thx!
Rich

Rich Jesse                           System/Database Administrator
[EMAIL PROTECTED]                  Quad/Tech Inc, Sussex, WI USA


> -----Original Message-----
> From: Freeman Robert - IL [mailto:[EMAIL PROTECTED]
> Sent: Friday, June 20, 2003 12:58 PM
> To: Jesse, Rich; ''Multiple recipients of list ORACLE-L ' '
> Subject: RE: What can cause boost in LIO/PIO?
>
>
> makes sence to me... I've seen an analyze, if it takes a
> long time to run,
> cause odd things to happen to SQL that starts running in the
> middle of the
> analyze process.
>
> The SAN rebuild certainly sounds like a likley suspect.
>
> RF
>
>
> -----Original Message-----
> From: Jesse, Rich
> To: Jesse, Rich; Freeman Robert - IL; 'Multiple recipients of
> list ORACLE-L
> '
> Sent: 6/20/2003 12:52 PM
> Subject: RE: What can cause boost in LIO/PIO?
>
> [deft backpedaling]
>
> OK, now I think it IS the SAN. The S/A handling the SAN is helping us
> with
> our backup scenario so he's reorganizing some drives on the SAN. At
> about
> the same time he started the reorg (highly drive-intensive), my
> aggregate
> PIO rate dropped from almost 10K/s to less than 2K/s. And the LIOs
> dropped
> similarly, while the event waits increased.
>
> I'm thinking that the impact of the reorg on the SAN caused the wait
> events
> for the PIOs to grow (from 3.5K/s to 4.7K/s). This longer
> wait in turn
> caused a decrease in LIOs because the processes are too busy
> waiting to
> generate the IO requests.
>
> This explains everything except that the lack of contention
> to cause the
> IOs
> to boost in the first place happened at the same time as the first run
> of
> the DBMS_JOB at 5:00 AM. Coincidence?
>
> How's that sound?
>
> Rich
>
> Rich Jesse System/Database Administrator
> [EMAIL PROTECTED] Quad/Tech Inc, Sussex, WI USA
>
>
> > -----Original Message-----
> > From: Jesse, Rich
> > Sent: Friday, June 20, 2003 12:32 PM
> > To: 'Freeman Robert - IL'; 'Multiple recipients of list ORACLE-L '
> > Subject: RE: What can cause boost in LIO/PIO?
> >
> >
> > Only in part. It's an hourly complete rebuild of a
> > mini-warehouse table (don't ask -- not my idea). It does do
> > a few index creates and accompanying analyzes of those
> > indexes and their table at the end of the procedure. I think
> > this accounts for the very high PIO spikes at the end of
> > every DBMS_JOB cycle, but I'm still not sure how it affected
> > the other queries.
> >
> > Thx!
> > Rich
> >
> > Rich Jesse System/Database Administrator
> > [EMAIL PROTECTED] Quad/Tech Inc, Sussex, WI USA
> >
> >
> > > -----Original Message-----
> > > From: Freeman Robert - IL [mailto:[EMAIL PROTECTED]
> > > Sent: Friday, June 20, 2003 11:06 AM
> > > To: Jesse, Rich; 'Multiple recipients of list ORACLE-L '
> > > Subject: RE: What can cause boost in LIO/PIO?
> > >
> > >
> > > What was this 5am job? It wasn't an analyze was it?
> > >
> > > RF
> > >
> > > -----Original Message-----
> > > From: Jesse, Rich
> > > To: Multiple recipients of list ORACLE-L
> > > Sent: 6/20/2003 11:29 AM
> > > Subject: What can cause boost in LIO/PIO?
> > >
> > > Hey all,
> > >
> > > I'm testing out the max I/O thruput of an IBM FastT900
> using a dual
> > > 2.4GHz
> > > w/1GB RAM on Win2K server (not my choice but it's just
> for testing)
> > > running
> > > Oracle 8.1.7. For one of my tests, I created the following
> > > heirarchical
> > > query:
> > >
> > > [yadda yadda yadda]
> >
>

-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.net
-- 
Author: Jesse, Rich
  INET: [EMAIL PROTECTED]

Fat City Network Services    -- 858-538-5051 http://www.fatcity.com
San Diego, California        -- Mailing list and web hosting services
---------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).
Received on Fri Jun 20 2003 - 13:46:08 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US