Path: news.daimi.aau.dk!news.uni-c.dk!sunic!sunic.sunet.se!news.funet.fi!news.csc.fi!news.eunet.fi!news.spb.su!KremlSun!satisfy.kiae.su!news.techno.ru!demos!uunet!panix!news.mathworks.com!udel!gatech!swrinde!cs.utexas.edu!uwm.edu!spool.mu.edu!olivea!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: 11 Apr 1995 21:51:36 GMT Organization: Sun Microsystems Inc., Mountain View, CA Lines: 131 Message-ID: <3metl8$h16@engnews2.Eng.Sun.COM> References: <3mdqej$6sd@fbi-news.informatik.uni-dortmund.de> 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) Thanks for your interest in BETA. I shall try to answer your questions with repects to the current support for concurrency and the plans for future support. With respect to experience with discrete event simulation, I am not able to give a full answer, but I know that the people at Mjolner Informatics have send out some enquiries to people who might have some experience with this. > as a computer science student at 'Universitat Dortmund' > I am writing a thesis in the field 'Parallel Discrete > Event Simulation (PDES)'. > > SIMULA has been applied with great success to 'Discrete > Event Simulation (DES)'. One of SIMULA's strength (with > respect to DES) is the concept of 'coroutines', but SIMULA > lacks concepts for 'concurrency'. > > I am wondering if BETA - being in the SIMULA tradition and > containing a powerful concept for concurrency - could be > applied to PDES with comparable success. > > Thus currently I am evaluating the BETA language: > > I am using the Mjolner BETA System version 5.0(2) for Linux > (Intel 386/486), Personal Edition ('Sonderlizenz fur den > Fachbereich Informatik, Universitat Dortmund'). > > This version does NOT support pre-emptive multitasking: > > "THE CONCURRENCY IS SIMULATED > In order to implement real concurreny, an interrupt > mechanism must be implemented. This is currently NOT done. > A component/system will thus keep the control until it > makes an explicit or implicit SUSPEND. An implicit SUSPEND > is made when a component must wait for a semaphore, executes > the pause pattern, executes the sleep pattern, or performs a > blocking communication using the shellEnv distribution > abstractions (not described in this manual). As the > concurrency is simulated, there is no difference between the > implementation of the alt and conc imperatives." > > [The Mjolner BETA System Basic Libraries Reference Manual, > MIA 90-08(1.2), August 1994, chapter 11, page 95 and preamble > of 'basicsystemenv.bet'] > > 1. Is there a version of BETA that supports real concurrency? The current implementation does NOT support real concurrency within one process. For the first implementations of BETA we decided to go for coroutine support a la Simula within one process. We have, however, implementation support for full concurrency through the Mjolner BETA distribution library. I.e. the Mjolner BETA System provides support for concurrent BETA programs in the form of distributed objects. The distribution mechanism has been used for some time, e.g. to implement the distributed hypermedia and object-oriented database described in Comm. ACM, Feb, 1994. You may also check the Mjolner BETA home page (see below) for further information. > 2. Will there be a version that supports real concurrency > in the near future? The reason for not including implementation of full concurrency in the first implementation was lack of resources. It has, however, very high priority to implement support for true concurrency within one process. > 3. Is there a version that supports multiprocessors (eg. for > SUN Workstations with Solaris 2)? The goal is that the support for true concurrency will include support for eg. SUN multiprocessors with Solaris 2. > 4. Will there be a version that supports multiprocessors in > the near future? Yes - but I dont know when it will be available, but we will keep you up-to-date with any news on this matter. > 5. Has anybody already applied BETA successfully to 'Discrete > Event Simulation'? I am not aware of anybody who has used BETA for practical discrete event simulation, although some customers have said that they planned to use it for that. I have for teaching purposes had students implement simulation packages a la class Simulation from Simula. And one Masters Thesis project in addition implemented support for graphical animation of such simulations. If you are intersted, a copy of that Thesis might be available from Computer Science Department, Aarhus University (I dont have the full reference here) > 6. Has anybody already applied BETA successfully to > '_PARALLEL_ Discrete Event Simulation'? > Again, I cannot give any concrete referencs, but some people have mentioned that they were interested in this. > Any hints and pointers to other sources of information are > welcome. Please eMail me: > Please let me know if you need further information. ---olm *************** BETA information Sources *********************************** WWW: http://www.mjolner.dk http://www.daimi.aau.dk/~beta/info News: comp.lang.beta FAQ: http://www.daimi.aau.dk/~beta/FAQ E-mail: info@mjolner.dk Address: Mjolner Informatics, Science Park Aarhus, Gustav Wieds Vej 10, DK-8000, Aarhus C, DENMARK Tel.: +45 86 20 20 00 Fax.: +45 86 20 12 22 ***************************************************************************** > Yours sincerely, > Armin M. Warda > ------------------------------------------------------------- > Armin M. Warda warda@ls4.informatik.uni-dortmund.de > Tel.: +49 (0)231 755-4824 (office) / +49 (0)231 753730 (home) > http://ls4-www.informatik.uni-dortmund.de/QM/MA/warda/pi.html > PGP fngrprnt 10 3B EC F5 BD 1E 83 7C 89 D1 9E 98 48 77 67 F7 >