Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!snorkelwacker.mit.edu!ai-lab!wookumz.gnu.ai.mit.edu!evil
From: evil@wookumz.gnu.ai.mit.edu (David Nye Evil)
Newsgroups: comp.benchmarks
Subject: Re: Upper bound on numbers of users
Message-ID: <14298@life.ai.mit.edu>
Date: 23 Mar 91 03:22:28 GMT
References: <1991Mar20.155013.18691@ima.isc.com> <14374@encore.Encore.COM>
Sender: news@ai.mit.edu
Organization: War Lord and Imperial Emperor of The Second Evil Empire
Lines: 30

In article <14374@encore.Encore.COM> soper@encore.UUCP (Pete Soper,Cary NC,9194813730,9193624635) writes:
>From article <1991Mar20.155013.18691@ima.isc.com>, by suitti@ima.isc.com (Stephen Uitti):
>> 
>> I've seen very good consistent user-response on multi-CPU
>> machines, such as from Sequent & Multimax.  With lots of CPUs, it
>> looked like you could run huge numbers of users.  I never saw
>> that many.
>
>  Multimax is the name of an Encore system. The model 320 I'm
>typing this on has 12 32332 cpus and has 168 users logged on at
>the moment; in the afternoon the number approaches 200. There are
>other systems within Encore that support 150-200 users and this
>does not count an unknown number of "invisible" xterm sessions.
>
	I recently got a chance to play on a Sequent with 20+ processors
Lots of memory, etc..and I swear to gawd...it had *576* users with a load
of .7!!!!  Most of the users were Xterm clients doing general things
but some of them were software types compiling etc.  This was mid-afternoon
usages.

All you can say to that is WOW...IMHO... :)

-=Dave Nye


--
-------------------------------------------------------------------------------
- Dave Nye,  a.k.a. -=TheEvilOne   -   evil@bbn.com   The Second Evil Empire  -
- AMIGA..need I say more?  }:)     -   evil@wookumz.gnu.ai.mit.edu            -
-------------------------------------------------------------------------------