ADSM-L

Re: Sizing of Bufpoolsize, everyone take note of this response

2003-01-20 09:47:38
Subject: Re: Sizing of Bufpoolsize, everyone take note of this response
From: PAC Brion Arnaud <Arnaud.Brion AT PANALPINA DOT COM>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Mon, 20 Jan 2003 15:46:48 +0100
Hi all (and especially Paul, Paul and Raphael !),

 Many thanks for your clear and very instructive responses : this is
going to be a precious help for defining better memory setting !
Cheers.

Arnaud

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
| Arnaud Brion, Panalpina Management Ltd., IT Group     |
| Viaduktstrasse 42, P.O. Box, 4002 Basel - Switzerland |
| Phone: +41 61 226 19 78 / Fax: +41 61 226 17 01       | 
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=



-----Original Message-----
From: Seay, Paul [mailto:seay_pd AT NAPTHEON DOT COM] 
Sent: Monday, 20 January, 2003 4:36
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: Sizing of Bufpoolsize, everyone take note of this response


We are about to upgrade to a 6 x 750 MHZ from a 4 x 450 MHZ.  At that
point, I will consider doing more bufferpool.  The problem is the buffer
processing does use some CPU resources, especially during expiration.
So, I will be able to model the benefits.  We are adding 2GB of memory
(total 4GB), so I may run my bufferpool up to a gigabyte.

Paul D. Seay, Jr.
Technical Specialist
Naptheon Inc.
757-688-8180


-----Original Message-----
From: Paul Ripke [mailto:stixpjr AT BIGPOND.NET DOT AU]
Sent: Friday, January 17, 2003 8:48 PM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: Sizing of Bufpoolsize, everyone take note of this response


On Saturday, January 18, 2003, at 02:36 AM, Seay, Paul wrote:

> As Zorg would say, I know the sound of this music.
>
> The default maxperm is probably killing you.  I am guessing you are 
> swapping
>
> <snip>

I agree whole-heartedly! I'd go a step further, and knock maxperm down
to about 10%, or even less. Given the nature of TSM I/O, the AIX buffer
cache is going to be next to useless. It may then be possible to expand
the TSM bufpoolsize beyond recommendations... I've set mine to about
60-70% physical RAM, up from 30-40%. The speed of selects and "q actlog"
have increased by an order of magnitude. Definitely stay below the level
where the system starts to swap, and don't go too large... TSM buffer
cache management can then become a bottleneck. How big is too large? No
idea - it'd be very dependent on hardware - experiment and see!

Cheers,
--
Paul Ripke
Unix/OpenVMS/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>