Group Title: Department of Computer and Information Science and Engineering Technical Reports
Title: Web-based simulation : some personal observations
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00095386/00001
 Material Information
Title: Web-based simulation : some personal observations
Series Title: Department of Computer and Information Science and Engineering Technical Reports
Physical Description: Book
Language: English
Creator: Fishwick, Paul A.
Affiliation: University of Florida
Publisher: Department of Computer and Information Science and Engineering, University of Florida
Place of Publication: Gainesville, Fla.
Copyright Date: 1996
 Record Information
Bibliographic ID: UF00095386
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.

Downloads

This item has the following downloads:

1996225 ( PDF )


Full Text








WEB-BASED SIMULATION: SOME PERSONAL OBSERVATIONS


Paul A. 1 1, !:

Dept. of Computer and Information Science and Engineering
University of Florida
Bldg. ('Si. Room 301
Gainesville, FL 32611


ABSTRACT

The web has a short history, but has grown at
an exponential pace since its introduction six years
ago. Web-based simulation represents the connection
between the web and the field of simulation. Web-
based simulation is not an existing field but rather
an idea which represents an interest on the part of
simulationists to exploit web I. i,. .1..- To further
this cause, I present some issues and concepts in web-
based simulation to serve as a back drop for a more
formal discussion, and potentially the formation of a
new simulation sub-area.

INTRODUCTION TO THE WEB

What is web-based simulation? I i! -1 we need to dis-
cuss a few terms before we can adequately address
this question. But, before doing that, we need to ,
something about the nature of this topic. The web
represents a rapid growth of Internet-based interac-
tion over the past six years. Yet, the topic reflects
more of a i 1... .l... 1,,-, ..... .,area than one with a
solid theoretic underpinning. Rapid advances have
taken place through practical application, engineer-
ing and entrepreneurial activity. A new web advance-
ment will appear, almost overnight, as a working prod-
uct to be tried by eager users. This situation-not
unlike the gold rush in the mid-nineteenth century-
suggests that, by default, an article on "--1. I,-. 1. I
-1in,! ,l,,i needs to focus on i .1 Yet, as re-
searchers, we collectively require a substantial knowl-
edge of this 1. !,. .1. -- if we are to apply the web
to our disciplines. It is not healthy to ignore the
web and remain fixated to either purely academic
tomes or -1. i- '- approaches to simulation. My
purpose in this article is to inform the reader about
some of the possibilities with an applications and
S1.1 1..!.- transfer focus. I have been involved as
a heavy user of web I. 1! .1,_ and will impart some
of ii personal experiences with the web as they re-


late to simulation. I will unabashedly refer to spe-
cific packages and methods, not with the deliber-
ate intention of presenting a biased view, but be-
cause this is not a survey paper and there are too
I iii web-based methods to be able to develop a
timely paper without making the content obsolete
once it is in print. Indeed, novel web-based applica-
tions seem to crop up so frequently that information
on web-based simulation will likely change as rapidly
as these applications. In this sense, the article is
somewhat anecdotal, and is meant to help simulation-
interested readers who are beginning to explore the
web, and possibly even simulation. Furthermore, in
the spirit of the -1, i. rI matter, all of our references
will be universal resource locators (URLs) and a hy-
permedia copy of this document should be accessed
to be able to link appropriately where required. The
hypermedia document is available by pointing your
browser at: http://www.cise.uf l.edu/~f ishwick
/websim.html as soon as you are near a PC or work-
station. B. -..i i this unique URL, which serves as
the gateway to all further citations and references,
all other URLs will not be visible in this article for
the sole reason that their ii .1.. iin,. i is not possible
without a browser. Nothing is gained from reading
the printed URL, and with a browser, you will not
require this document. So, I encourage you at the
next opp I,,l u-t to go to a browser at locate the pre-
vious URL. It is assumed that you will have access to
a browser, as there are several free browsers available
and some (such as Microsoft's Internet Navigator)
come bundled with their operating -I. i i- (\\ ,i 1... -
95 and NT), which in turn are bundled with all PCs.
The PC is the most popular platform for web-based
applications due to the huge number of PCs and the
PC open architecture. Other supported platforms in-
clude Apple's MacIntosh and Unix workstations such
as the ones made by Sun, Hewlett-Packard and Sili-
con Graphics. Most web software is currently pack-
aged for \\ i! .l..- 95 first, then \\ !!!...;- NT and
\\ 1, ... - 3.1, followed by Unix versions.









The Internet is a loosely connected world-wide
network of heterogeneous computers. The World \\ ,1
Web (WWW) is a set of on-line hypermedia docu-
ments that reside on the Internet. Hypermedia doc-
uments are built using a language called hypertext
markup language (HTML). HTML started as a sub-
set of S( ;., I (lS ,I 1!1 11 1, Graphics Markup Language)
but grew to absorb the needs of web authors. Ac-
cess to the web is afforded using a web browser such
as Netscape, Mosaic or Internet Navigator. When
you use the browser, you are presented with a de-
fault home page which presents you with your first
online document. This document will include /t,,- -
links which, when clicked using a mouse, link to an-
other document. In this way, a document is connected
to other documents in a sprawling network: there is
no centralized '.. !!ii i ,-1 documents loosely con-
nected to one another through links. Documents i, ,
contain more than the usual printed document:

Text: various natural languages are supported.

Images: I 1i'. 111 GIF tIl. seem to be the best
supported but JPEG is also supported. Image
converters exist for most platforms.

Video: supported video I I," are MPEG and
AVI, with the additional GIF89 that supports
inline small video clips meant for small icon dis-


Audio: AU and WAV are supported formats, as
well as digital standards such as MIDI.

Although web browsers support 11ii 1 exciting fea-
tures, our purpose is to something about how the
web and the field of simulation relate to one another:
web-bases simulation. Given a browser and the abil-
ity to write new web pages, what are the issues for
simulationists? How can we exploit the new technolo-
gies and how will our field change as a result of the
web?

HOW SIMULATION WILL BE AFFECTED

There are !! ii aspects to simulation as there are
for i,!! scientific discipline. We will focus on the fol-
lowing aspects: 1) Education and Training, 2) Pub-
lications, 3) Simulation Programs. All three of these
topics are interrelated. For the academic readers,
the topics of education and publication are probably
on the top of the list, and industrial simulationists
will want to focus more on training and the practical
methods for building simulations that use the web.


Education and Training

How do we currently educate our students using sim-
ulation? Most modern simulation packages f ..1,
have graphical front ends (so called graphical user
interfaces-GUIs). This allows a student to learn
through virtual world exploration. The web will af-
fect this process in several ways. I ,-, a diskette
or CD-ROM has a limited amount of local storage
but the web provides us with effectively infinite stor-
age. The "--*I ,i.-" is on the Internet and not limited
to one's own machine. Therefore simulation pack-
ages which include help text and information about
the pieces of the model, can have these pieces on
the web so that they do not require local storage.
This encourages some collaboration in that a com-
,i marketing their simulation tool, conveniently
priced for educators, will have to think more globally
about information and resources. It is not necessary
for a 'ii | ii to re-build every piece of information
about a manufacturing process, for instance, from
scratch. Information on automated guided vehicles
(AGVs), machine specifications, and automated con-
, 1,,,,. mechanisms ii ,- already be located some-
where on the web. These sorts of devices are common
in simulation programs built for manufacturing sys-
tems i ,1 -- Why re-invent the wheel by building
our own information databases on these manufactur-
ing devices? The web encourages this sort of global
view with a re-use of knowledge and information.
While academics ii revel in this kind of reusable
knowledge, industry will have to rethink certain pro-
prietary concerns: a *-,! iii,- needs to make a profit
to survive. They also need to produce iiL 1i- simula-
tion software, but this does not negate the use of the
information of other web sites. Credit must be given
where it is due, as usual, and it is not clear exactly
how **I' 1- i1 trademarks and patents will change
as a result of the web phenomenon. It is too early to
tell, but we need to address these issues now.
We have discussed the use of web sites in user doc-
umentation for a simulation program, but the pro-
gram could just as easily be placed on the web as
well. After all, is the program really that itl! i. i I
from hypermedia "- I .. II- !iI Ii. !i '" We discuss this
in more detail within this section. The most profound
affect the web i i have on our method of teaching
simulation lies with the use of multimedia, since it
is multimedia that we find in every web link. The
web encourages distance learning more so than the
!- I1,' 1 simulation textbook. On o!i- web page it is
possible to include images and video of the instruc-
tor along with i1 !! .111 ,- 1 slides or overheads. This
immerses the student in a ii 11, i learning environ-
ment that is more congenial than one they would get









simply by reading a book or watching a videotape.
Various technologies are around to aid us in creating
multimedia web pages for simulation education. Be-
fore mentioning a few of them, it is worth defining
the concepts of client and server since the web en-
capsulates these two roles. A server provides a client
with information and services. The daily metaphors
of servers and clients (whether going to a restaurant
or visiting your doctor or lawyer) are the best way of
understanding these roles. If you enter your browser
and access a travel ;,-. in Connecticut, you are
able to do this because the travel :,- i. has a com-
puter which contains a web server. This web server
services your needs by delivering multimedia docu-
ments to you via your browser. The relationship is
directional: the;,: --. is the server and you (or your
machine) are the client. Every machine, potentially,
has the ;,1.l i- to be both server and client so there
is no reason why the ;,.- in might not use your ma-
chine as a server if you have it set up as such, deliv-
ering information required by the travel -,. in Any
machine can serve, then, as both client and server.
A plug-in is a software program to allow third
l' i' vendors (with respect to the browser designer)
to provide multimedia service for web pages. For ex-
ample, if I install the PDF reader from Adobe, Inc.
into Netscape, I will then be able to include PDF
I l. in one of ii- web pages, so when I access some-
one elses web page which contains a server ', .I_1,1i *
delivering PDF til. i!- PDF plug-in will activate
and I will be able to see PDF til. Plug-ins exist for
numerous functions. A recent trend has been for com-
mercial vendors to provide plug-ins for free to ;o! i.
but to charge a fee for those wanting to serve clients.
If you want, for example, to design your own PDF
! l. so that others can access them over the web, this
will involve a cost to you, and this is where the com-
' i" generates its required revenue. If web software
is free, it is free only in the sense that someone else is
I 1 ;- for you to use it in the same way that TV and
Radio advertisers 1 for the programming which you
, i ,
What plug-ins are useful to a simulation educator?
When you have a need to deliver real-time video and
audio to your students, then you will need to purchase
the server software, while your students can freely ob-
tain the client-side plug-in. For non-real time video
and audio, !1 iii, free 1i1 '- exist (e.g., MPEG and
AVI for video and AU, MIDI for audio), and so most
companies seem to be making a profit for providing a
much-needed real-time audio-video stream, especially
over low bandwidth connections such as via 2- -i
modems. VDOLIVE and VDOPHONE are especially
good at providing close-to real time audio and video.
VDOLIVE is a method of -1i. 11,11i video and au-


dio in the same way that a TV does. VDOPHONE
is an Internet substitute for a videophone. Both of
these packages i !i.1.- wavelet compression to achieve
excellent performance. Video frame rate is generally
sacrificed where necessary to maintain uninterrupted
audio. This is as it should be since ",-1.1!:. iC' video
is acceptable, but this is not true for audio. Occa-
sionally, another marketing technique is to provide
a "I" p -'"'! -- i for free. This provides you with
the :,l1ili- to use one stream, which is fine as long as
two clients (i.e., students) do not attempt to access
the video stream simultaneously. !1 I other plugins
are useful as well, such as one that can display color
slides, possibly in i i ir ,i with the audio and
video streaming plugins.
In the sense that our conferences represent ". 1-
ucation and training," we need to make 11i i..r! over-
hauls of the way we do business. The use of tele-
conferencing will help us to interact prior to meet-
ing at a conference. Personal meetings with each
other cannot be replaced by electronic means, but
new web tools such as CU-I. .., 11. and VDOPHONE
(ref. Publications) have the means to allow us to
communicate more effectively. E-mail, mailing lists
and i S 1.' I news groups (including the new web-
based news groups) are useful technologies. The i Si.' I
news group comp. simulation has been in operation
for almost a decade and has helped to grow the disci-
pline. The ELE( 'S.l conference has been success-
ful in forging ahead to permit us to communicate
with one another without forcing cross-country trips
to conferences.

Publications

We publish for one reason: to inform others of new
or existing simulation research and applications. We
write articles in -. l!ii, from conference proceed-
ings such as this one, and trade magazines to aca-
demic archival journals. The web is dramatically
changing the way we publish, and it is changing so
fast that iii i- are rethinking how publications are
delivered and purchased. Let's consider archival sim-
ulation journals first. If you publish an article in an
archival journal, there is an accepted practice which
is defined as follows:

1. P, ,f.. '. you submit a manuscript to the
journal's Editor in ('I!. I The editor then has
an editorial board which is used to manage the
review process for the paper. The idea is that,
for your paper to be accepted, it should be sent
to qualified reviewers who are experts in the
sub-simulation area that represents your paper's
focus. Moreover, these reviewers are ;!!..!,-
mous so as not to bias the review. Sometimes,









conferences try to snip part of an author's sub-
mission so that the reviewer doesn't know who
wrote the article. This procedure tends not to
be practiced in the simulation niiiii,~L~ and
the procedure has flaws since self-reference (to
establish background) in a paper generally links
a specific person to the paper. Once the pa-
per is reviewed, it can be i. i. 1 I1 accepted or
-. i. J I to revision with provided comments and
critique by the reviewers. A review process from
time of submission to a decision can take Ii, -
where from three months to six months.

2. Publication: The review process can be 1. -1 ,i ,
and assuming that you have made the proper
revisions to your paper, your paper i i be ac-
cepted by the Editor in ('I!. on recommenda-
tion of a member of the board. The time from
acceptance to the time your paper appears de-
pends on two things: the periodicity of the jour-
nal and the current queue (backlog) of already
accepted papers. Times from six months to two
years are common.

3. Archival: Libraries I';' i11 archive journals so
that others can search for your article (using a
scientific index) for their own research.

This entire process is 1. ,ii -! but there is good rea-
son for much of the time taken to evaluate the paper.
We, as scientists, also accept this method since it has
proven to be the way to introduce original simula-
tion knowledge to grow our information base. The
web has changed the I'I i1- field 1 !!! new ideas
are surfacing. Most of the simulation magazines and
journals are making new inroads here. For example,
both the ACM Transactions on Modeling and Com-
puter Simulation (TOI l.\ 'S) and the SCS Transac-
tions on Simulation are rapidly transitioning to elec-
tronic form. Let's consider some issues:
1. Literature Search: a publication begins with
a thorough literature search. This search is
made easier on the web through search tools,
but searching for the right information is still
lI!,I 111i There is so much information that
searches can become unwieldy. Often, searches
on the web result in "! -! 11 link" problems: all
information is distributed and so links change
dynamically. Only evolved and maintained links
remain in way not unlike the processes of ge-
netic variation and fitness in living organisms.
Libraries are changing their functions to help
with information access and retrieval. There
is still an art to locating just the right infor-
mation and this is the task of an information
specialist-the modern librarian.


2. !./, : ..... Access: make i 11 electronic.
The author submits the paper using an elec-
tronic medium such as Postscript, which is fast
becoming a kind of pseudo-standard for page
1 ., Li among journals. Referees are sent (either
via mail attachments or using I!!.- transfer-ftp)
a compressed Postscript tid. Referees submit
their reviews by e-mail and the author is in-
formed of the decision by e-mail as well. All
of this makes sense and is happening now. All
citations use URL links instead of printing a
'.1i.1,- '1,1! '. This allows the reader to jump
directly to a cited paper, avoiding 1. ,ii -! and
sometimes unsuccessful literature searches.

3. 1'P, ,... '.,f more comments can be made avail-
able to an author if !ii. .i!- is permitted to at-
tach comments to an article being considered
for publication. Why limit the comments to
two or three people? We call this "--. ,1. i refer-
.. ii, : let those who are interested in reading
an article make comments on it. One possibil-
ity is to have a combination of traditional and
reader refereeing.

4. Revenue: who i' - for the article? Before, ev-
l!ii_ was so simple the. !1ii which pub-
lished the hardcopy of the journal received basi-
cally all the revenue. However, if the volunteers
(editorial board) are doing all the work and do
not need the publisher, then should simulation
journals be free? Well, there are costs lurking
everywhere: the cost of the server equipment
and the cost of archiving journal articles. Some-
one has to i' for these costs. Costs have been
translated from paper-printing into digital ser-
vice and storage.

5. Preprints: remember that the goal of publish-
ing is for you to impart information to others
on your new ideas. If you can put your papers
immediately online in your web page, and have
these web pages indexed by web robots and in-
telligent agents, then you don't have to wait, iii
longer to see your work "in print." It is in print
immediately. The generally accepted practice
is for authors to place preprints on the web, ei-
ther in a home page or in a preprint area main-
tained by someone else, and to simultaneously
submit the article for consideration by a jour-
nal. Whereas before, only a handful of people
might ever thoroughly read your journal arti-
cle, now : ii I i -- with a web browser can find it
using a search engine and read it. Yet, there is
so much information on the web, that we want
to narrow our search. As societies abandon pa-









per in favor of web-publishing, there still will
be value in doing your search for simulation pa-
pers via society servers first since that is where
papers have been checked for ','i l,-I

Simulation Programs

The most exciting part of web-based simulation is in
the simulations themselves. Operating on the web
gives new meaning to the word document. A web
document can contain videos, interactive simulations,
images and audio in addition to the usual text that
traditional documents contain. How do we embed
simulations in documents? A simulation requires a
model and a model requires computing hardware to
execute the model. Some parts of a model can be
executed on another machine and other parts are ex-
ecuted on your machine where your web browser is
located. For -i!i ~1.Ii i we'll assume that you are the
one building the simulation and others want to ex-
ecute it. To build a simulation which executes on
your machine requires that someone else (the client)
access your simulation (the server) and cause a pro-
gram to execute. Typically scripting languages such
as Perl are used to build a high level program that
runs your simulation code once you have obtained all
required simulation inputs and parameters, generally
using a ''"!t- i i i.' l.1 If 1 ,1ii users are run-
ning your simulation simultaneously, you will need to
make sure your Perl script is set up to handle this.
Your computer resources 1! be taxed, depending
on the number of readers wanting to run the simu-
lation. The idea of Java and Javascript is to bundle
a piece of code (called an apple) and send it to the
client's computer to execute, not yours. This way,
it makes no difference how i 1, i!~ clients want to run
your simulation at least from the perspective of model
execution.
In our discussion so far, you have constructed a
simulation which runs on your computer or sends ap-
plets to a client's computer. This is fine for single-user
simulations, but what if you need to have a simulation
where multiple users interact with each other? The
best example of user interaction is found in multi-user
gaming simulations such as DOOM and QUAKE. In
such a simulation, a global server (one person's ma-
chine) is used to capture and maintain the state of
the world. Each I.1 i makes moves and informa-
tion is moved from client to server. Multi-User Dun-
geons (MUD) operate in a similar fashion, the dif-
ference being that the interaction among users 1ii'
extend toward a cooperative relation to other 1 i' -
rather than a strictly competitive one. The I li 1) is
also a basis for 1, i I education through simulation
within a virtual simulated "-'I 1 This represents a


shared solution to multi-user simulation.
If the simulation model is spread over a network
of computers, then we have a distributed solution.
Each computer is associated with either a single 1i1! --
ical. i I i or a set of entities. The distributed inter-
active simulation (DIS) initiative begun by the De-
partment of Defense (DoD) is a good example of dis-
tributed model execution. The DoD focus is to have
an interactive simulation 'I. l1,.li- mainly for train-
ing people, but increasingly other aspects of the mil-
itary (such as ";~ ili-II i' I ) are finding uses for DIS.
Some of the computers in a DIS network 11 i be run-
ning semi-automated forces (SAFOR) and others will
have humans attached to often sophisticated human-
computer interfaces so that the training exercise is
made as realistic as possible. If there is no real-
time human-computer interaction (HCI) necessary,
we iii still split our models and do parallel and dis-
tributed simulation (using conservative or optimistic
time advance).
How does the web relate to all of this? Since the
web is a collection of documents, an interactive simu-
lation could ostensibly be entered from within a web
document. A user locates a document, presses the
mouse button on an image pi 1, i 11 a virtual world
and then is immersed in a simulation. Most DIS sim-
ulations do not currently operate in this fashion, but
there is no reason why this could not be done. Af-
ter all, the web is all about storing and retrieving
multimedia information, and simulation is a key kind
of "'., ..... ,"',-. The virtual !. ,-i i markup language
(VI:. 1L) suggests a way to do this. VI:. 11 plug-
ins allow one to construct 3D geometry and have it
rendered in real time. A document contains a URL
which points to a VI:. 11 I. td The browser recog-
nizes this particular information I- ,. and launches
the VI:. 11. plugin. Then the user browses a 3D scene
using whatever HCI devices are available.
\\ ,I I! DIS and I- 1i )s, we are only scratching the
surface of what is possible. You have probably no-
ticed that most companies currently have web sites.
What if they were to take the products that they
manufacture and put digital equivalents of these prod-
ucts on their web site? This would radically change
the ;- ii which we do modeling and simulation. One
of the biggest reasons why simulation of large -- -1 i i -
is expensive is that each one of us rebuilds model com-
ponents instead of re-using what could be out there
on the web. Customers will benefit from such web-
based models since they can try out a product before
purchasing it. This also goes for industrial customers
as much as it does for individuals. For the original
equipment manufacturers, this scheme will allow for
them to sell more equipment since potential buyers
can try out the equipment simply by linking the part









of their simulation model which models the equip-
ment to the manufacturer's web site. Manufacturers
will have to spend more time building digital web-
based models for their products, but this will be time
well spent once readers and users locate the models.
We have done some preliminary studies of web-
based simulation at the Uni- i -i of Florida. Fig. 1
displays the home web page for a simulation of four
disks and one CPU. A job enters the CPU and then
is assigned a disk unit for ti!.- access. Then, this job
loops a certain number of times. This looping simu-
lates the time taken by an average program (i.e., job)
which has need of CPU and tihI- storage resources.
The web page is organized in an overall linear fash-
ion, starting with a paper title, author list and then
proceeding with a picture of the -- -1. i to be simu-
lated. The square area of the screen uses an "IS., I.\ P"
feature which allows us (the authors of the web page)
to perform a given action if the reader clicks on a
certain region within this square. At the bottom of
Fig. 1, the user clicks on this to begin entering sim-
ulation input information. All input is entered using
the FOi;. IS feature of HTML. Once the the infor-
mation has been entered, the reader can execute the
simulation. Since our simulation uses the "-- i i side
approach," using a Perl script which invokes an ex-
ecutable on a SPARC Unix workstation, the simu-
lation executes and produces time-dependent graphs
for CPU and Disk resource queue sizes. The graphs
are produced by dynamically creating an HTML web
page from the simulation output, which is then avail-
able to the reader.

REDUCING THE COST OF SIMULATION

Simulation has always been a costly enterprise, and
even though more inexpensive computer equipment is
driving costs down, the ever-lurking cost makes itself
known in all of our concerns: education, publication
and simulation programs. We have just discussed re-
ducing this cost by the re-use of objects within sim-
ulation models but there is a broader -I11 1. _- ad-
vertising. F.- .- i!. has questions about how !i -ii
is to be effectively exchanged over the web. Rev-
enue issues will affect the simulation .' iiin iiil ii as
much as ;, ii other, so we need to consider the trans-
fer of iii. ii. Simulation products and services can
be bought through traditional credit card means, and
we have seen that 1, i~ services (in the form of web
products) are free for the client side of information
delivery. The cost is on the server side. I! i simu-
lation services 11i be possible through inventive ad-
vertising methods such as including an advertiser's
message in a web page accessed by simulation practi-
tioners. Most web-search engines contain this feature.


The web user benefits by using the search engine and
the search engine service is paid for by the adver-
tiser. F. i. ..!- wins. In this way, education, publi-
cation and simulation software 11i i become cheaper
as a result of embedded advertisements. We should
be concerned about over-commercializing simulation
knowledge; however, web advertisements are less ob-
trusive to the reader and provide direct links to the
advertiser. Furthermore, by the web's nature, adver-
tiser information tends to be more information-rich
compared to other I- i" of media such as television,
radio and magazines. The advertising links are also
interactive so there is the potential for providing qual-
ity information to readers. This means that simu-
lation publications, educational products and inter-
active simulation programs of the near future might
incorporate direct advertising to cover costs. The ad-
vertising is less blatant, more accommodating, and
provides for a three-way benefit to the server, client
and advertiser.

SUMMARY

We are in a kind of C-. I,,- -I i ,- century gold rush
since i ..i- is rushing to put multimedia informa-
tion on the web, but there is great uncertainty about
how things such look and feel. How will simulation-
related societies change and deliver their publications?
Will companies expand their sales by offering simu-
lation models of their products? One thing is for
certain. We will never know :ii1 of these answers un-
less we experiment and take risks if necessary. The
"ioiji way to do -ii ,l Ii.i will naturally emerge
and evolve as we try various web-based simulation ap-
proaches. The worst possible approach is to sit back
and adopt a "I. I v- oi attitude: simulation will be
worse off as a discipline unless we move forward now
to incorporate the web-based technologies.

ACKNOWLEDGMENTS

I would like to acknowledge the following funding
sources which have contributed towards our study
of modeling and simulation. (1) Rome Laboratory,
(; ,it!-- Air Force Base, New York under contract
I ;1i 1 1'-95-C-0267and grant I ; 11 '--95-1-0031; (2) De-
partment of the Interior under grant 14-45-0009-1544-
154 and the (3) National Science Foundation Engi-
neering Research Center (i.i i1) in Particle Science
and T !!!i. .1. .- at the U! i.- i-. of Florida (with
Industrial Partners of the 1. il1) under grant EEC-
94-02989.

AUTHOR BIOGRAPHY












An Interactive Web Simulation of CPU/Disk Performance

Paul Fishwi M Matthew Belk, Bad1eye pt S

We present a functional model of a CPU/Disk subsystem with an interactive execution using SimPack.
Aside from providing technical data on the model, the paper serves as an experiment m publishing an
interactive simulation on the World Wide Web. See Citing and Referencing Web Publications for
further information. Let's consider a combined CPU/Disk system (Aai2 lt:yod u :
Point to a feature for an explanation. V" nfefined
(job cycle) fyperfins

S te toI Simula te
s stem to Simute
Il l V I I -- I I I^ "


- Z lL ULi Y tiLC--L! ~-- ISMaP rRegion
-u.E&ecutte Per!fScript (Run Simuation)


1 ,-i .- 1: Web page of a CPU/Disk simulation.











Paul A. Fishwick is an Associate Professor in the
Department of Computer and Information Science
and Engineering at the University of Florida. He re-
ceived the PhD in Computer and Information Sci-
ence from the University of P. !i!--1- i in 1986.
He also has six years of industrial/government pro-
duction and research experience working at Newport
News !,llIilil, and Dry Dock Co. (doing CAD/
CAM parts definition research) and at NASA Lang-
ley Research Center (-l.ll- i![_ engineering data base
models for structural. i! -I11!. _. His research inter-
ests are in computer simulation modeling and anal-
ysis methods for complex -1. 1i- He is a senior
member of the IEEE and the Society for Computer
Simulation. He is also a member of the IEEE So-
ciety for Systems, Man and Cybernetics, ACM and
AAAI. Dr. I i-l!- ,i !: founded the comp.simulation
Internet news group (Simulation Digest) in 1987. He
has chaired workshops and conferences in the area
of computer simulation, and will serve as General
('I! i! of the 2000 \\ !!!. i Simulation Conference. He
was chairman of the IEEE Computer Society techni-
cal committee on simulation (T( 'Si. !) for two years
(1988-1990) and he is on the editorial boards of sev-
eral journals including the .I 1 .1 .... ... ... on Mod-
eling and Computer Simulation, IEEE 1. ...I ...
on R !. ...*. M an and C ;., .., ',. I! I...... ...
of the Society for Computer Simulation, International
Journal of Computer Simulation, and the Journal of
.,/ ...., Engineering. Dr. I -!-! 1 !1: WWW home
page is http://www.cis.ufl.edu/ ~fishwick and
his E-mail address is fishwick@cis.ufl. edu.




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs