RE: more processors = increased runtime ?

["Glenn Birnbaum" <gbirnba@xxxxxxx>]

Even though the 12 way model may in total CPW be a larger system than
the 4 way, it really matters what the speed of an individual processor
is rated at.  For example, I have a model 820/2435 single processor that
is rated at 600 cpw.  I have a 720/2063 dual processor rated at 810 cpw,
but each processor would only be 405 cpw.  So it wouldn't surprise me if
the 'smaller' single processor 820 out performed the 'larger' dual
processor 720.  (note that SMP would change this since a job could run
on more than one processor).

Try dividing the rated CPW by the number of processors and see what the
result would be on each system for a single processor.

HTH,
Glenn Birnbaum

-----Original Message-----
From: Jim Damato [mailto:jdamato@dollargeneral.com]
Sent: Tuesday, January 15, 2002 10:56 AM
To: 'midrange-l@midrange.com'
Subject: RE: more processors = increased runtime ?


>  Nigel:
>For one program that is called within the process, the elapsed run time

>is 1.5hrs on System A whereas it is 8.0 hrs on System B.

>System A is a 4-way system, system B is a 12-way system. Both are on
>the
same
>version of OS/400 with the same CUM PTF's applied. The subsystem
descriptions
>should be similar and both run in the base pool.

What are the system features/models?  Since this is a single
program/process it's going to boil down to i/o throughput and processor
speed.  Is the 12-way using older CPU's than the 4-way?

-Jim

James P. Damato
Manager - Technical Administration
Dollar General Corporation
<mailto:jdamato@dollargeneral.com>
_______________________________________________
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing
list To post a message email: MIDRANGE-L@midrange.com To subscribe,
unsubscribe, or change list options,
visit: http://lists.midrange.com/cgi-bin/listinfo/midrange-l
or email: MIDRANGE-L-request@midrange.com
Before posting, please take a moment to review the archives
at http://archive.midrange.com/midrange-l.