Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!know!samsung!uakari.primate.wisc.edu!sdd.hp.com!elroy.jpl.nasa.gov!ucla-cs!frazier@oahu.cs.ucla.edu
From: frazier@oahu.cs.ucla.edu (Greg Frazier)
Newsgroups: comp.realtime
Subject: Help with references, distributed realtime systems
Message-ID: <38232@shemp.CS.UCLA.EDU>
Date: 21 Aug 90 18:51:36 GMT
Sender: news@CS.UCLA.EDU
Organization: UCLA Computer Science Department
Lines: 26

Hello.
	I need some help either confirming or correcting
my impressions, hopefully with some quotable reference
for backup.
	Under simple real-time systems, one ensures that
every operation can occur within a bounded time period,
thus guarenteeing to meet external deadlines.  For complex
systems, particularly distributed ones, this is not possible.
A probabalistic approach is taken in these situations.  For
example, I am currently working on a communications network.
When implementing a realtime system upon it, one would want
to be able to say "at a network throughput of 30%, a packet
will reach its destination in less than 90 time units with
a prabability of 0.9999."
	I would appreciate hearing from designers of complex
(hopefully distributed) realtime systems who have or have
not taken this approach, and I really would like references
to conference proceedings or journal articles which discuss
this.  Also, I understand that this is called "soft" realtime.
A reference for that (or any) name for this would also be
appreciated.  Post or email.
--
"They thought to use and shame me but I win out by nature, because a true
freak cannot be made.  A true freak must be born."  K. Dunn, _Geek_Love_

Greg Frazier	frazier@CS.UCLA.EDU	!{ucbvax,rutgers}!ucla-cs!frazier