HP3000-L Archives

May 1996, Week 5

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Stan Sieler <[log in to unmask]>
Reply To:
Stan Sieler <[log in to unmask]>
Date:
Wed, 29 May 1996 11:41:09 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (39 lines)
Bryan asks:
> We run our own home grown data collection program using the MI.  We collect
> "heaps" of statistics and the job only uses about 1,000 CPU seconds each 24
> hours.
...
> I am a little unsure if it is the MI job which is supposed to put a load on
> the system or if it stresses the other processes.
 
Easy enough to test...take the sample from my posting last week (?), and
try it: with and without having your (or any) MI-based program running.
 
A simple test, although it accents CM code, is:
 
turn off (or on) MI, and then:
 
   :fcopy from=nl.pub.sys;to=$null
   (discard first timing)
 
   :fcopy from=nl.pub.sys;to=$null
   (measure with stopwatch)
 
turn on (or off) MI, and then;
 
   :fcopy from=nl.pub.sys;to=$null
   (discard first timing)
 
   :fcopy from=nl.pub.sys;to=$null
   (measure with stopwatch)
 
You discard the first timing each time in order to minimize the hidden
difference of how many pages might be in memory through luck.  In this
manner, about the same number of pages will be in memory when you start
the second (timed) test.
 
 
--
Stan Sieler                                          [log in to unmask]
                                     http://www.allegro.com/sieler.html

ATOM RSS1 RSS2