ADSM-L

Re: AIX and async I/O--additional info

2003-04-24 17:38:18
Subject: Re: AIX and async I/O--additional info
From: Paul Ripke <stixpjr AT BIGPOND.NET DOT AU>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Fri, 25 Apr 2003 07:37:50 +1000
On Friday, Apr 25, 2003, at 06:39 Australia/Sydney, PINNI, BALANAND
(SBCSI) wrote:

Paul I am not from AUS ,rather from USA as shown in below message .

Sorry - that's just my mail clients default behaviour...

Paul as u are DBA did u try any time working with disk_asyncio in
init.ora .
I am not talking about aio that is used while running rootpre.sh that
creates /etc/pwsyscall .

No - my main dealings with Oracle and AIO have been regarding some
unlogged AIO failures to HDS, on AIX 4.3.3. The ticket with IBM was
closed due to insufficient evidence, and our inability to reproduce
the problem.

Just looked at the Oracle doco - disk_asynch_io is true by default. And
on some of our systems, we'd definitely need it that way. Setting it
to false would serialise writes which would bog down the DB writer.

BTW, I believe Oracle on AIX only uses AIO for writes - the DB writer
flushing dirty blocks. Interesting to watch a system take ~2 minutes
for disk I/O to idle down after an Oracle dump.

<snip>

Cheers,
--
Paul Ripke
Unix/OpenVMS/TSM/DBA
101 reasons why you can't find your Sysadmin:
68: It's 9AM. He/She is not working that late.
-- Koos van den Hout

<Prev in Thread] Current Thread [Next in Thread>