Path: news.daimi.aau.dk!news.uni-c.dk!sunic!sunic.sunet.se!news.kth.se!nac.no!Norway.EU.net!EU.net!news.sprintlink.net!cs.utexas.edu!venus.sun.com!news2me.EBay.Sun.COM!engnews2.Eng.Sun.COM!olm
From: olm@Eng.Sun.COM (Ole Lehrmann Madsen)
Newsgroups: comp.lang.beta
Subject: Re: PDES / Concurrency in the BETA Language
Date: 14 Apr 1995 21:24:16 GMT
Organization: Sun Microsystems Inc., Mountain View, CA
Lines: 27
Message-ID: <3mmp60$dof@engnews2.Eng.Sun.COM>
References: <3mdqej$6sd@fbi-news.informatik.uni-dortmund.de> <3metl8$h16@engnews2.Eng.Sun.COM> <3mgb73$dn8@harbinger.cc.monash.edu.au>
NNTP-Posting-Host: det.eng.sun.com
Keywords: PDES,parallel,discrete,event,simulation,Mjolner,BETA,concurrency
X-Newsreader: NN version 6.5.0 #21 (NOV)

dcmay1@mdw082.cc.monash.edu.au (Mr DC May) writes:
>What is the granularity of the concurrency that BETA will support?
>Particularly, I am referring to the difference between monolithic
>Unix-type processes and lightweight threads (e.g. OS/2).

I dont know much about OS/2.
The current BETA implementation has a form of leightweight processes
already. Objects of kind component are coroutines in the style of
Simula. The Mjolner BETA System has a framework for supporting
various forms of abstractions on top of components. These
include monitors, Ada-like tasks, etc. See 

	$BETALIB/basiclib/v1.4/systemenv.bet

The current implementation only has non-preempty scheduling of
components. We are working on a preemptive version of this. I.e.
the components will be like leightweight threads within one thread.

As said in my previous posting true concurrency can be obtained
through the distribution framework.

---olm

>	Daniel


>P.S.  Did you get the echidna I left for you?  :)