Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site ucbvax.ARPA
Path: utzoo!watmath!clyde!burl!ulysses!ucbvax!info-vax
From: info-vax@ucbvax.ARPA
Newsgroups: fa.info-vax
Subject: Real-time/clusters
Message-ID: <4091@ucbvax.ARPA>
Date: Wed, 9-Jan-85 13:45:47 EST
Article-I.D.: ucbvax.4091
Posted: Wed Jan  9 13:45:47 1985
Date-Received: Fri, 11-Jan-85 08:10:13 EST
Sender: daemon@ucbvax.ARPA
Organization: University of California at Berkeley
Lines: 30

From: MSE20800%TIMEVX%CITHEX@lbl.arpa

Received: from timevx by cithex with DECNET ; Wed, 9 Jan 85 10:43:16 PST
X-ST-Return-Receipt-Requested:
Date:     Wed, 9 Jan 85 10:43:41 PST
From:     mse20800 (ewing, martin s.) @ timevx
Message-Id: <850109104341.004@timevx>
Subject:  Real-time/clusters
To:       INFO-VAX @ SRI-CSL.ARPA
 
 
 
I am new to info-vax, and a little overwhelmed.  However, I have a real
question:
 
How many have used Clusters, and how many have attempted real-time work
with Clusters?  We are planning a system that might benefit from using
clustered disks and tape.  One VAX would run in a pretty strict real-time
regimen controlling a hardware processor, and one would run in a more lax
manner massaging output data on its way to tape.
 
The problem is possible contention slowing the real-time machine.  Would
anyone offer an opinion?  Is it realistic to put any system files out on
the cluster?
 
Thanks for your attention.
 
Martin S. Ewing ( MSE @ CALTECH.ARPA)
Caltech 105-24
Pasadena, CA 91125