Subject: | |
From: | |
Reply To: | |
Date: | Wed, 6 Feb 2008 15:04:00 -0500 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
[log in to unmask] wrote:
> I'm doing an analysis to possibly upgrade our production 959KS/100 system to a 979KS/200 and I see the performance metric chart that tells me we go from a 4.6 to 14.6. Does anyone have or know where I can get a general idea what the increase represents? for instance, does each whole number (like 4.0 to 5.0) represent a general percentage increase in performance?
It is a relative number to a base system model. The number varies
according to the list of systems.
No documentation that I know of has ever been published regarding any
relative performance benchmarks.
Thus, YMMV,Objects May Be Closer Than They Appear ... and all other
disclaimers apply.
> I know it varies from one shop to another, so I'm just looking for a general guideline or personal experience like a job that used to take 10 hours to run now only takes 7 hours. The "personal experience" part of this may not even be appropriate, in that the upgrades may not be close to the metrics I am looking at.
Unless the processes within the job are written to take advantage of the
multiple processors (like TurboStore/iX) a single
stand alone job won't execute in any shorter time than the single CPU
version. Thus your 979ks/200 would really only be
a /100 to that job. In real life few HP3000 batch workloads are anywhere
near %100 CPU. A look at the CPU/CONNECT
times in the stdlist is a good rough estimate of CPU %. Again if it is a
stand alone job. A good procedure is to insert showme
commands between job steps to get the split times for individual job steps.
In my experience, specifically investigating the longest job steps can
be very rewarding in terms of performance gains.
Mike Hornsby
CTO/CoFounder
Beechglen Development Inc.
513-922-0509x31
* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
|
|
|