simulation@uflorida.cis.ufl.edu (Moderator: Paul Fishwick) (03/17/91)
Archive-name: internet/simulator/lansf/1991-03-15
Archive: menaik.cs.ualberta.ca:/pub/lansf.2.11e.tar.Z [129.128.4.241]
Original-posting-by: simulation@uflorida.cis.ufl.edu (Moderator: Paul Fishwick)
Original-subject: Network simulators from SIMULATION DIGEST V20 N10
Reposted-by: emv@msen.com (Edward Vielmetti, MSEN)
Volume: 20, Issue: 10, Fri Mar 15 17:02:30 EST 1991
TODAY'S TOPICS [edited for comp.archives. --Ed.]
(3) FREE: Communication Network Simulators
* 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.
------------------------------
From: pawel@cs.UAlberta.CA (Pawel Gburzynski)
Subject: Network simulation tools (ethernet models in particular)
To: simulation@bikini.cis.ufl.edu
Date: Mon, 11 Mar 91 16:07:39 MST
X-Mailer: ELM [version 2.3 PL11]
I would like to announce two free simulation tools that are available
from our site. A while ago somebody inquired about an ethernet model; well,
the ethernet model is a trivial addition to these tools.
The tools I am talking about are LANSF (Software Practice and
Experience, vol. 21, no. 1, January 1991, pp. 51-76) and SMURPH (see the
Proceedings of 1991 Western Multi Conference on Object-Oriented Simulation,
Simulation Series, vol. 23, no.3, 1991, pp. 12-21). The second package is
brand new and no serious papers on it have been published yet. Below I
include a random excerpt from a technical report on SMURPH.
----
SMURPH is an object-oriented software package for modeling discrete events
in communication protocols. It can be viewed as both an implementation of a
certain protocol specification language and a process-driven simulator
oriented towards investigating medium access control (MAC) level protocols.
The structure of the simulator is completely invisible to the user.
All the simulation-related operations, as creating and scheduling
individual events, maintaining a consistent notion of time, etc.,
are covered by a high-level interface. A protocol description in SMURPH
looks like a program that could be executed on a hypothetical hardware.
SMURPH emulates this hardware and thus provides a realistic environment
for executing protocols described in its specification language.
Protocol execution in SMURPH can be monitored. One reason for monitoring the
behavior of a protocol is to investigate its performance by gathering empirical
data. Although SMURPH does not purport to be a protocol verification system
(it seems that no automated verification system of practical use can be built
today), it offers some tools for protocol testing. These tools, the so-called
observers, look like programmable dynamic assertions describing legitimate
sequences of protocol actions.
The protocol description language of SMURPH is essentially C++.
The standard data types, objects, functions, and macros provided by the
package extend this language substantially. With this approach, the user
gets the full power of C++ combined with the power of a realistic, emulated
environment for programming and executing communication protocols.
SMURPH descends from its predecessor, called LANSF which was designed and
implemented in 1987. LANSF was programmed in C and has evolved a number of
times from its original version. It has been successfully applied to
investigating the performance and correctness of a number of protocols for
local and metropolitan area networks and other distributed systems.
One disadvantage of LANSF was its low flexibility in describing compound data
structures. The idea of re-implementing LANSF in C++ was born from the
collaboration of the researchers from UofA with the networking group at the
Lockheed Space and Missile Company. At first, Objective C was tried, but
after a brief preliminary design we decided to use C++ instead. C++ seems to
capture all the essential features of Objective C, and at the same time is
devoid of some deficiencies of that language.
----
Both packages are available for free although we try to solicit a token and
voluntary contribution from the users -- to partially cover the maintenance
cost. SMURPH is still in a development stage, and I will welcome all
suggestions, bug reports, etc. It currently works with the GNU C++ compiler,
but I have already started adapting it to the AT&T cfront environment.
The packages are available from menaik.cs.ualberta.ca (129.128.4.241), files
"pub/lansf..." and "pub/smurph...", where "..." depends on the current
version of the product. They come with extensive documentation and examples
(Ethernet is among them).
For the people who have no ftp access, we can arrange for some other means of
transportation.
Below is the list of selected relevant papers (in the bibtex format):
----
@article( ber89
,author= {Bertan, B. R.}
,title= {Simulation of {MAC} Layer Queueing and Priority Strategies of {CEBus}}
,journal= {IEEE Trans. on Consumer Electronics}
,volume= 35
,month= aug
,year= 1989
,pages= {557--563}
)
@inproceedings( dog88
,author= {Dobosiewicz, W. and Gburzy\'{n}ski, P.}
,title= {Ethernet with Segmented Carrier}
,booktitle= {Proceedings of IEEE Computer Networking Symposium}
,year= 1988
,month= apr
,address= {Washington, DC}
,pages= {72--78}
)
@inproceedings( dog90
,author= {Dobosiewicz, W. and Gburzy\'{n}ski, P.}
,title= {Performance of {P}iggyback {E}thernet}
,booktitle= {IEEE IPCCC'90}
,year= 1990
,month= mar
,address= {Scottsdale, AZ}
,pages= {516--522}
)
@inproceedings( dgr90
,author= {Dobosiewicz, W. and Gburzy\'{n}ski, P. and Rudnicki, P.}
,title= {An {E}thernet-like {CSMA/CD} Protocol for High Speed Bus {LAN}s}
,booktitle= {IEEE INFOCOM'90}
,year= 1990
,pages= {238--245}
)
@techreport( gbm90
,author= {Gburzynski, P. and Maitan, J.}
,title= {An object-oriented configurable simulator for low-level communication protocols}
,institution= {University of Alberta}
,address= {Edmonton, Alberta, Canada}
,type= {Technical Report TR 90-8}
,month= mar
,year= 1980
)
@inproceedings( gbr87
,author= {Gburzy\'{n}ski, P. and Rudnicki, P.}
,title= {A Better-than-{T}oken Protocol with Bounded Packet Delay Time for Ethernet-type {LAN}'s}
,booktitle= {Symposium on the Simulation of Computer Networks}
,publisher= {IEEE}
,year= 1987
,month= aug
,address= {Colorado Springs, Co.}
,pages= {110--117}
)
@inproceedings( gbr88a
,author= {Gburzy\'{n}ski, P. and Rudnicki, P.}
,title= {Modeling Low-level Communication Protocols: A Modular Approach}
,booktitle= {Proceedings of the 4th International Conference on Modeling Techniques and Tools for Computer Performance Evaluation}
,publisher= {Universitat de les Illes Balears}
,year= 1988
,address= {Palma de Mallorca, Spain}
,pages= {89--108}
)
@article( gbr88c
,author= {Gburzy\'{n}ski, P. and Rudnicki, P.}
,title= {A Virtual Token Protocol for Bus Networks: Correctness and Performance}
,journal= {INFOR}
,volume= 27
,year= 1989
,pages= {183--205}
)
@inproceedings( gbr89
,author= {Gburzy\'{n}ski, P. and Rudnicki, P.}
,title= {On Formal Modelling of Communication Channels}
,booktitle= {IEEE INFOCOM'89}
,year= 1989
,pages= {143--151}
)
@article( gbr89a
,author= {Gburzy\'{n}ski, P and Rudnicki, P.}
,title= {A Note on the Performance of {ENET II}}
,journal= {IEEE Journal on Selected Areas in Communications}
,month= apr
,year= 1989
,volume= 7
,pages= {424--427}
)
@manual( gbr89b
,author= {Gburzy\'{n}ski, P. and Rudnicki, P.}
,title= {The {LANSF} Protocol Modeling Environment, version 2.0}
,address= {The University of Alberta, Department of Computing Science, TR 89-19, Edmonton}
,year= 1989
)
@inproceedings( gbr89c
,author= {Gburzy\'{n}ski, P. and Rudnicki, P.}
,title= {On Executable Specifications, Validation, and Testing of {MAC}-level Protocols}
,booktitle= {Proceedings of the 9th IFIP WG 6.1 Int. Symposium on Protocol Specification, Testing, and Verification, June 1989, Enschede, The Netherlands}
,editor= {Brinksma, E. and Scollo, G. and Vissers C. A.}
,publisher= {North-Holland}
,year= 1990
,pages= {261--273}
)
@incollection( gbr89d
,author= {Gburzy\'{n}ski, P. and Rudnicki, P.}
,title= {{LANSF}---A Modular System for Modelling Low-level communication protocols}
,booktitle= {Modeling Techniques and Tools for Computer Performance Evaluation}
,editor= {Puigjaner, R. and Potier, D.}
,publisher= {Plenum Publishing Company}
,year= 1989
,pages= {77--93}
)
@inproceedings( gbr90
,author= {Gburzy\'{n}ski, P. and Rudnicki, P.}
,title= {Modelling of reactive systems in {SMURPH}}
,booktitle= {Proceedings of the European Simulation Multiconference}
,address= {Erlangen-Nuremberg, W-Germany}
,month= jun
,year= 1990
,pages= {661--667}
)
@article( gbr91a
,author= {Gburzy\'{n}ski, P. and Rudnicki, P.}
,title= {{LANSF:} A Protocol Modelling Environment and its Implementation}
,journal= {Software Practice and Experience}
,volume= 21
,number= 1
,year= 1991
,month= jan
,pages= {51--76}
)
@inproceedings( gbr91
,author= {Gburzy\'{n}ski, P. and Rudnicki, P.}
,title= {Object-Oriented Simulation is {SMURPH}: A Case Study of DQDB Protocol}
,booktitle= {Proceedings of 1991 Western Multi Conference on Object-Oriented Simulation}
,address= {Anaheim, California}
,month= feb
,year= 1991
,pages= {12--21}
)
@article( pam90
,author= {Pakkam, S. R. and Manikopoulos, C. N.}
,title= {Performance Evaluation of the {C}onsumer {E}lectronic {B}us}
,journal= {IEEE Transactions on Computer Electronics}
,year= 1990
,note= {(In press.)}
)
----
My best regards,
Pawel Gburzynski
Department of Computing Science
University of Alberta
615 GSB
Edmonton, Alberta, CANADA T6G 2H1
tel.: (403) 492-2347
fax.: (403) 492-1071
e-mail: pawel@cs.ualberta.ca
------------------------------