simulation@uflorida.cis.ufl.edu (Moderator: Paul Fishwick) (12/13/90)
Volume: 19, Issue: 3, Wed Dec 12 20:50:24 EST 1990 +----------------+ | TODAY'S TOPICS | +----------------+ (1) RE: Recent Inquiries (2) CALL: Workshop on Computer Aided Systems Theory (3) Modelling and Simulating LANs (4) SPECIAL: Simulation Tools (A Review) * Moderator: Paul Fishwick, Univ. of Florida * Send topical mail to: simulation@bikini.cis.ufl.edu OR post to comp.simulation via USENET * Archives available via FTP to bikini.cis.ufl.edu (128.227.224.1). Login as 'ftp', use your last name as the password, change directory to pub/simdigest. Do 'type binary' before any file xfers. * Simulation Tools available by doing above and changing the directory to pub/simdigest/tools. ----------------------------------------------------------------------------- Date: Wed, 5 Dec 90 08:41:02 EST From: Richard E. Nance - SRC <srcnance@popeye.cs.vt.edu> To: simulation@bikini.cis.ufl.edu Subject: Recent Inquiries Cc: janlee@vtopus.cs.vt.edu Heavens, do I feel old when people ask questions like "What is the language dynamo (sic) like and how do I get it?" and "Does anyone know anything about the history of Simscript and if it is still used?" Bet Joe Annino at CACI loved that last one. Pugh-Roberts & Associates (I believe) still distributes Dynamo, although whether it is Dynamo II or III or ? is not in my knowledge base. The second question does suggest to me the need to document the history of Simscript. We have good documentation on Simula and GPSS (History of Programming Languages, Academic Press, 1980) but the only historical description of Simscript, as I recall, is in Jean Sammet's book on programming languages. Someone should capture this material while Markowitz, Kiviat and others are still around and capable of contributing. Maybe this is a task of interest to Annals of the History of Computing. Dick Nance ------------------------------ Date: Wed, 5 Dec 1990 13:38 MST From: ZEIGLER%EVAX2@Arizona.edu Subject: Eurocast To: fishwick@fish.cis.ufl.edu X-Envelope-To: fishwick@fish.cis.ufl.edu X-Vms-To: PAUL Call for papers: Eurocast '91: Second International Workshop on Computer Aided Systems Theory, Krems Austria, Aprill 15-19, 1991. Send 3 copies of an extended abstract before Jan. 20, 1991 to Prof. Franz Pichler, Institute of Systems Science, Kepler University, A-4040 Linz, Austria. Fax: +43 732 2468 896, email: k323290@aearn.bitnet. ------------------------------ Date: Thu, 6 Dec 90 15:58:29 EST From: rkimble@NSWC-WO.ARPA (Robert Kimble) To: fishwick@fish.cis.ufl.edu Subject: LAN Modelling Software I am interested in purchasing a software package which is geared towards modelling and simulating local area networks and I'd like to get some suggestions as to what is currently available that would best satisfy my requirements and perhaps some pros/cons of the competing packages. The basic requirements are: The package should be able to model the effects of using different networking media and networked devices. The package should have pre-defined (or "canned") all of the necessary characteristics required for accurate modelling of industry-standardized or very popular network media and devices. In particular FDDI, IEEE 802.x, IBM Token Ring as well as point-to-point medias would be required. Network devices would include IBM XT and AT class computers using 3COM ethernet cards, Sun Microsystems workstations, and 3COM and ciscoSystems bridges and routers. The package should also allow for changing the values of those pre-defined characteristics to enable modelling of non-industry-stasndardized or popular equipment. The networks to be modelled will be using standard networking protocols (mainly just three different ones: TCP/IP, IPX, and GOSIP). Are there any modelling packages that model the transport layer protocols of these three networking protocols? Or are the effect of the transport layer protols not significant. I would think that the flow control (windowing) mechanisms of the transport layer would have a predominant effect especially when slow devices such as PCs are receiving from faster devices such as workstations. Any comments or suggestions would be appreciated. Robert Kimble rkimble@nswc-wo.navy.mil 301-394-4225 ------------------------------ To: uunet!comp-simulation@uunet.UU.NET Path: motcid!reilly From: motcid!reilly@uunet.UU.NET (Patrick L. Reilly) Newsgroups: comp.simulation,local.tools.local Subject: Re: SIMSCRIPT(& Tools for You!) Date: 7 Dec 90 20:39:39 GMT References: <25708@uflorida.cis.ufl.EDU> Reply-To: motcid!reillyp@uunet.UU.NET Organization: Motorola Inc., Cellular Infrastructure Div., Arlington Heights, IL simulation@uflorida.cis.ufl.edu writes: :I am curious about the historical background of SIMSCRIPT. :Does anyone have any info? :Also is it still a popular language for simulation? :If not, what is? Was developed by Markowitz (current Nobel Prize Winner for economics) and others at Rand Corporation during 60s. Still very popular language. The government uses it alot. Some 440K lines of code used in a SDI project. Also several battle management tools exist. I use it as my primary language to code complex simulations. It has all the power of FORTRAN and/or C and then some. That power comes at a price...CPU performance is about 2-3 times slower than FORTRAN or C. Any simulationist worth the name should have a suite of simulation tools. Here are mine(means that I USE them, too)... 1. Simscript II.5 for code-level programs of any complexity 2. SES/Workbench for graphical simulation modeling 3. Simpak, public domain implementation of SIMPL 4. Mathematica 5. GPSS/H student version (100 blocks only! But what a challenge!) 6. QSB+ 7. PROBMOD/LINPROG 8. COMNET II.5 I run all of this and more on a Sun 3/470 color workstation. From time to time I also use NCSA CRAYs, too. Simscript is available on the CRAYS as well, but I have not used it there. Ok, here is a run down of the items mentioned: 1. CACI, Inc. LaJolla, CA. Call 619-457-9681 $20-50K Worth every penny if you are an applied scientist like myself and not interested in the subtleties of computer science. I have not found anything I could not accomplish with it. Also includes a graphical simulation animator interface to build user screens, icons, menus, etc. 2. SES, Inc., Austin, Tx. Call 512-474-4526 $26-45K Nothing currently available comes close to this product. It is a graphical environment by the company that brought us the PAWS simulator. Lots of object-oriented extensions. Just connect icons from libraries and run. Build your own with method blocks using C-language. Planning a Software thru Pictures interface 1st qtr 91. Also VHDL planned. Can simulate a VLSI chip or an ethernet. Company is still learning how to deal with large customers, so you will have to educate them about your particular needs sometimes. I predict that this tool will be the standard by which others will be judged. 3. Available from this newsgroup ftp site. FREE! If you just getting started in simulation find TWO books: -Simulation Modeling & Analysis, Law/Kelton, McGraw-Hill, 1990 -Simulating Computer Systems, MacDougall, MIT Press, 1987 The first book, now in 2nd Edition, has it all. Even the source simulation tool code in Fortran is included. This is Simulation 101, folks. My bible! The second book describes a C-implemention of SIMPL, a basic simulation tool used by the author for everything from queueing to ethernets. This newsgroup has had several postings about Dr. Fishwick's SIMPAK tool. Consider it a more refined version of SIMPL and much more. Get it!!! [I leave it to the moderator to include the ftp information.] 4. Mathematica needs no introduction. Nor any comments. Right? Call any computer store for a copy. 5. Wolverine Software Corp., Call 703-750-3910 $40!!! Get their student version and a book by Banks, Getting Started Using GPSS/H for a trivial sum. Comes with disk for PC. Limits you to 100 GPSS/H blocks. Full PC and workstation versions about $3K-20K. This is NOT the old GPSS of the IBM flavor. It is FAST! Old GPSS was an interpreter, GPSS/H is not. I love the challenge of seeing how complex a model I can build with only 100 blocks. Sooner or later, I will spring for the full version. Be prepared, though. It is not like any language you have encountered, unless you are used to op codes and assembly language! 6. From Prentice-Hall Publishers. $35!!!! Quantitative Systems for Business Plus is a book with software for PCs aimed at the beginning operations researcher. Lots of nifty algorithms implemented in a friendly interface. I use it for quick solutions to queueing theory questions, linear programming,transportation problems,network/critical path modeling,PERT,dynamic programming,inventory theory, decision and probability theory,Markov processes and time series forecasting. It even comes with a queueing simulator! Call P-H or visit a local university bookstore in the operations research book section. 7. From McGraw-Hill. $65!!!! These programs accompany Hillier/Lieberman's ever-popular book, Introduction to Operations Research. It is also available in a two-volume set. The BEST BOOK ON THE SUBJECT! Hillier's son, Mark, wrote the programs and they run on a PC or Mac, so get the correct version. I use these as much as I use QSB+. There are differences and I mainly wanted the book. 8. Available from CACI, Inc. $25-35K Written in about 50K lines of Simscript with a graphical user interface. A programming-free communications network simulation tool. Circuit,packet,virtual,message-switching supported. Lots of the usual standards, too. I can whip out a C7 signaling network simulation in a hour or less with it so I like it! Haven't seen anything else out there that compares to it. This just shows you how powerful the Simscript language is, folks. Buy the compiler, write a tool, sell it and get rich! That about does it. Yes, I missed your favorite package. Well I happen to not be fond of OPNET,CSIM,SIMULA,GPSS/PC,SLAMII,etc. So there! Good luck and hope this helps you get started. ------------------------------ END OF SIMULATION DIGEST ************************