RE: %ASP Used

There are those that argue that 80% fell out of favor as a rule of thumb 
when they upgraded from a 5MB disk to a 10MB disk.  I've heard it when we 
had a S/36 at the high end that it no longer applied.  However I'd 
disagree.  I still think it is very applicable.  Yes there are lots of 
other considerations, such as disk arms, etc.  And, as one person 
mentioned, the overall size of your system.  For example we had a small 
system that would be 80% full and yet you still couldn't do a RGZPFM on 
the biggest file on the system - had to do it to tape.
The reason I think the 80% still applies is that you still have the same 
level of sporadic jumps, etc on the larger system as you did on the 
smaller system.  For example, as we grew the number of objects grew 
tremendously.  User's mail files, Tivoli Storage Manager's backups of 
their PC's, etc.  So instead of being just concerned about that "one file" 
having a spike, or needing the space to reorg, you have these thousands of 
objects that may fluctuate.  Thus the 80% rule still applies.

There are those that argue that your percent used of CPU should be 
approaching 100% on a regular basis.  These people have one thing in 
common - lousy response time.  Their argument is that the iSeries will use 
all that is available to do it's processing.  However our main production 
box has great response time unless we go over our interactive governor.
% CPU used . . . . . . . :       14.0

The disk on our main production box is fuller than we like.  But we are in 
the process of ordering more:
System ASP . . . . . . . :     2479 G
% system ASP used  . . . :    85.8166

We have 145 drives, ranging from 8gb to 35gb.  All raid 5 protected.

Our main development box:
% CPU used . . . . . . . :       79.0
System ASP . . . . . . :    337.3 G
% system ASP used  . . :    80.0692
We have 42 drives ranging from 6607, 6714, 6718, 6719


Rob Berendt
-- 
"They that can give up essential liberty to obtain a little temporary 
safety deserve neither liberty nor safety." 
Benjamin Franklin 




"Graap, Ken" <keg@xxxxxxxxxxxxx> 
Sent by: midrange-l-bounces@xxxxxxxxxxxx
11/19/2003 12:30 PM
Please respond to
Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxx>


To
"'Midrange Systems Technical Discussion'" <midrange-l@xxxxxxxxxxxx>
cc

Subject
RE: %ASP Used






>I like to run no more than 60% on average, and no more 
>than 80% under any conditions.

I don't know if this is as relevant anymore... On a terabyte system that
would mean having 200GB's of empty disk, at 80% utilization...

I think you need to understand what your disk IOP and disk arm utilization
is along with you disk usage growth history. If you have enough arms and
IOP's and the arm utilization is 5-8% then running in the 90% full range
should be acceptable. 

Another benifit would be not having to explain to a "bean counter" why the
iSeries won't run correctly unless it has 200GB of empty disk space!

Kenneth

****************************************
Kenneth E. Graap
IBM Certified Specialist 
AS/400e Professional System Administrator
NW Natural (Gas Services)
keg@xxxxxxxxxxxxx
Phone: 503-226-4211 x5537
FAX:    603-849-0591
****************************************




-----Original Message-----
From: michaelr_41@xxxxxxxxxxxxxx [mailto:michaelr_41@xxxxxxxxxxxxxx]
Sent: Wednesday, November 19, 2003 8:04 AM
To: Midrange Systems Technical Discussion
Subject: Re: %ASP Used

_______________________________________________
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing 
list
To post a message email: MIDRANGE-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/midrange-l.