@ARTICLE{Shin93:Mixed, AUTHOR="Kang G. Shin and Qin Zheng", TITLE="Mixed time-constrained and non-time-constrained communications in local area networks", JOURNAL=ieeecom, YEAR=1993, KEYWORDS="local area networks; medium access protocols; time constrained traffic; real-time traffic; token ring; token bus; FDDI; 802.4", ABSTRACT="It is well known that some timed-token medium access protocols for local area networks (LANs) like the IEEE 802.4 token bus and the FDDI token ring can guarantee the medium access delay for time constrained packets. However, a problem which has been largely overlooked is how these protocols can be made to provide a maximum throughput for non-time-constrained packets while guaranteeing the delay bound of time-constrained packets. We first show how the parameters of the IEEE 802.4 token bus and the FDDI token ring can be set to solve the above problem. Then, we design a new timer mechanism for the timed-token protocols which provides the highest guaranteed throughput of non-time-constrained packets among a set of medium access protocols called the token passing protocol, to which most the existing non-contention LAN protocols belong. We present numerical examples to compare different protocols, all of which have shown the superiority of the proposed protocol to others.", } @INPROCEEDINGS{Yaro9303:Calculating, AUTHOR="O. Yaron and M. Sidi", TITLE="Calculating Performance Bounds in Communication Networks", BOOKTITLE=infocom, ORGANIZATION="IEEE", ADDRESS="San Francisco, California", MONTH="March/April", YEAR=1993, KEYWORDS="bounds; quality of service; end-to-end analysis", } @INPROCEEDINGS{Sohr9303:Real, AUTHOR="K. Sohraby and A. E. Eckberg", TITLE="Real-time high percentile tracking of quasi-exponential delays", BOOKTITLE=infocom, ORGANIZATION="IEEE", ADDRESS="San Francisco, California", VOLUME=2, PAGES="850-857 (7c.4)", MONTH="March/April", YEAR=1993, KEYWORDS="delay estimation; tail estimation", ABSTRACT="In many applications certain parameters of a system should be set based on some large percentiles of a delay random variable, for which the characteristics may not be known, or may change slowly with time. Assuming that sample value of delays are available sequentially, we obtain a simple algorithm to track two key statistical characteristics of the delay. Any high percentile may then be approximated as a simple linear function of these parameters. The approach is based on a ``quasi-exponential'' assumption concerning the delays.", ANNOTE="might be useful for setting playout delay", } @INPROCEEDINGS{Sohr9303:Theory, AUTHOR="K. Sohraby", TITLE="On the theory of general on-off sources with applications in high-speed networks", BOOKTITLE=infocom, ORGANIZATION="IEEE", ADDRESS="San Francisco, California", VOLUME=2, PAGES="401-410 (4a.3)", MONTH="March/April", YEAR=1993, KEYWORDS="on-off sources; call admission; performance evaluation; statistical multiplexer", ABSTRACT="In this paper, we first provide a general theory of on-off sources. Our basic source model is characterized by alternating independent on (burst) and off (silence) periods, where these periods may have general distributions. Other more complex sources are constructed and their basic behavior is characterized in terms of the basic source model. Heterogeneous and homogeneous statistical multiplexers fed by such sources are considered. In the heterogeneous environment, a simple result on the tail behavior of the multiplexer queue length distribution in the heavy traffic is provided which is a simple function of the individual source peak rates, and the first two moments of their on and off periods. In the homogeneous environment, asymptotic results on the tail behavior of the queue length distribution is provided for all level of utilizations. In this case, the behavior strongly depends on the complete distribution of the on and off periods of the traffic source. The simple result in the heterogeneous environment, suggests a new call admission policy for general on-off sources in high speed networks. This policy can be easily implemented in real time, depends only on the first two moments of the on and off periods of individual sources and their respective peak rates.", } @INPROCEEDINGS{Rubi9303:Burst, AUTHOR="I. Rubin and Kuang-Shin Lin", TITLE="A burst-level adaptive input-rate flow control scheme for {ATM} networks", BOOKTITLE=infocom, ORGANIZATION="IEEE", ADDRESS="San Francisco, California", VOLUME=2, PAGES="386-394 (4a.1)", MONTH="March/April", YEAR=1993, KEYWORDS="flow control; rate control; multiplexer; ATM", ABSTRACT="We propose a new input rate flow control scheme wherein the credit increment rate is updated periodically as the loading state varies. Based upon the observed status of each station's burst-level activity, the network access node distributes feedback control signaling messages to the stations. These signaling messages allow the stations to adapt their credit increment rates in accordance with system burst-loading conditions. We present queueing models to study the system performance at the access points of such a system. For this purpose, we select a subnetwork topology which involves a network switch (such as a fast packet switch in high-speed metropolitan or wide area networks) and a number of regulated source stations which drive the network switch. To avoid packet retransmission due to cell losses at the access switch, each user station (or CPN) implements a local replica of the input regulation scheme. The output traffic streams from the source stations, as regulated by the local input rate control mechanism (and adapted by the status messages) load a packet switch which is modeled as a multiple-server queueing system. Performance curves are presented to illustrate the statistical queue-size behavior and message delays at both the source stations and the network switch.", } @MISC{ISO93:10646, AUTHOR="{International Standards Organization}", TITLE="{ISO/IEC} {DIS} 10646-1:1993 Information Technology -- Universal Multiple-Octet Coded Character Set {(UCS)} -- Part {I:} Architecture and Basic Multilingual Plane", YEAR=1993, KEYWORDS="character set; 10646", } @INPROCEEDINGS{Pare9303:Generalized, AUTHOR="Abhay Parekh and Robert Gallager", TITLE="A generalized processor sharing approach to flow control in integrated services networks -- the multiple node case", BOOKTITLE=infocom, ORGANIZATION="IEEE", ADDRESS="San Francisco, California", VOLUME=2, PAGES="521-530 (5a.1)", MONTH="March/April", YEAR=1993, KEYWORDS="delay bounds; weighted fair queuing; generalized processor sharing", ABSTRACT="Worst-case bounds on delay and backlog are derived for leaky bucket constrained sessions in arbitrary topology networks of generalized processor sharing (GPS) servers. When only a subset of the sessions are leaky-bucket constrained, we give succinct per-session bounds that are independent of the behavior of the other sessions and also of the network topology. However, these bounds are only shown to hold for each session that is guaranteed a backlog clearing rate that exceeds the token arrival rate of its leaky bucket. When all the sessions are leaky bucket constrained, a much larger class of networks called consistent relative session treatment (CRST) networks is analyzed. The session $i$ route is treated as a whole, yielding tighter bounds than those that result from adding the worst-case delays (backlogs) at each of the servers in the route. The bounds on delay and backlog for each session are efficiently computed from a universal service curve, and it is shown that these bounds are achieved by ``staggered'' greedy regimes when an independent session relaxation holds. Propagation delay is also incorporated into the model. Finally, the analysis of arbitrary topology GPS networks is related to packet GPS networks (PGPS).", } @INPROCEEDINGS{Doar9303:Bad, AUTHOR="Matthew Doar and I. Leslie", TITLE="How bad is naïve multicast routing?", BOOKTITLE=infocom, ORGANIZATION="IEEE", ADDRESS="San Francisco, California", VOLUME=1, PAGES="82-89 (1c.2)", MONTH="March/April", YEAR=1993, KEYWORDS="multicast; routing; Steiner tree; random network; random graphs", ABSTRACT="When the problem of routing multicast connections in networks has been previously considered, the emphasis has been on the source transmitting to a fixed set of destinations (the multicast group). There are some applications where destinations will join and leave the multicast group. Under these conditions, computing an ``optimal'' spanning tree after each modification may not be the best way to proceed. An alternative is to make modest alterations to an existing spanning tree to derive a new one. An extreme, though non-optimal, variation of this is to use minimal cost source to destination routing for each destination, effectively ignoring the existing multicast tree. We examine just how non-optimal these trees are in random general topology networks and conclude that they are worse by only a small factor. The factor is reduced still further if a hierarchy is imposed upon the random network to give a more realistic model.", } @ARTICLE{Arav9301:Image, AUTHOR="R. Aravind and Glenn L. Cash and D. L. Duttweiler and H.-M. Hang and B. G. Haskell and A. Puri", TITLE="Image and video coding standards", JOURNAL=atttj, VOLUME=71, NUMBER=1, PAGES="67-89", MONTH="January/February", YEAR=1993, KEYWORDS="MPEG; JPEG; JBIG; Px64; image compression; discrete cosine transform; packet video; DCT; video coding", ABSTRACT="This paper describes several standard compression algorithms developed in recent years.", } @BOOK{Come93:Internetworking, AUTHOR="D. Comer and David L. Stevens", TITLE="Internetworking with {TCP/IP} -- Client-Server Programming and Applications", ISBN="0-13-474222-2", PUBLISHER="Prentice Hall", ADDRESS="Englewood Cliffs, New Jersey", VOLUME=3, YEAR=1993, KEYWORDS="TCP; IP; protocol implementation; operating systems", } @BOOK{Pisc93:Open, AUTHOR="D. Piscitello and A. Lyman Chapin", TITLE="Open systems networking: {TCP/IP} and {OSI}", ISBN="0-201-56334-7", PUBLISHER="Addison-Wesley", ADDRESS="Reading, Massachusetts", YEAR=1993, KEYWORDS="OSI; TCP; IP; computer networks; network architecture; network standards", } @BOOK{Part93:Gigabit, AUTHOR="C. Partridge", TITLE="Gigabit networking", ISBN="0-201-56333-9", PUBLISHER="Addison-Wesley", ADDRESS="Reading, Massachusetts", YEAR=1993, KEYWORDS="computer networks; protocol architecture; survey", } @BOOK{Lync93:Internet, AUTHOR="Dan Lynch and Marshall T. Rose", TITLE="{Internet} system handbook", ISBN="0-201-56741-5", PUBLISHER="Addison-Wesley", ADDRESS="Reading, Massachusetts", YEAR=1993, KEYWORDS="Internet; TCP; IP; internetworking; survey; security", } @ARTICLE{Lin9304:HAP, AUTHOR="Y. Lin and Tsong-Ming Tsai and Shan-shan Huang and M. Gerla", TITLE="{HAP:} A New Model for Packet Arrivals", JOURNAL=jhsn, VOLUME=2, NUMBER=4, YEAR=1993, } @ARTICLE{Park9304:Warp, AUTHOR="Kun I. Park", TITLE="Warp control: a dynamically stable congestion protocol and its analysis", JOURNAL=jhsn, VOLUME=2, NUMBER=4, YEAR=1993, } @ARTICLE{Krau93:FCC, AUTHOR="Jeffery A. Krauss", TITLE="{FCC} creates new wireless business opportunities", JOURNAL="Applied Microwave and Wireless", PAGES="13-29", MONTH="Fall", YEAR=1993, KEYWORDS="FCC; spectrum allocation; PCS; wireless; cellular radio; mobile communications", } @INPROCEEDINGS{Hosc9303:Tailoring, AUTHOR="Philipp Hoschka", TITLE="Towards tailoring protocols to application specific requirements", BOOKTITLE=infocom, ORGANIZATION="IEEE", ADDRESS="San Francisco, California", PAGES="647-653 (5d.4.1)", MONTH="March/April", YEAR=1993, KEYWORDS="ASN.1; GOSIP; protocol implementation; protocol configuration", ABSTRACT="A distributed application that uses a standard protocol for communication very often applies only a subset of the protocol options provided by the standard. However, it is difficult today for an application developer to tailor standard protocol options to the specific needs of an application, because no appropriate profiling method (formalized profile description and protocol configuration tool) exists. In this paper, we describe a prototype of such a method that is applicable to any protocol specified in ASN.1. We use the profiling of OSI application layer standards as a proof of concept. By applying our method to formalize the U.S. government profiles we were able to reduce the source code size of major parts of common OSI applications by 30 to 50\%.", } @TECHREPORT{Hwan93:MDP, AUTHOR="RenHung Hwang and James F. Kurose and Donald F. Towsley", TITLE="{MDP} routing in {ATM} networks using the virtual path concept", TYPE="Technical Report", INSTITUTION="Department of Computer Science, University of Massachusetts", ADDRESS="Amherst, Massachusetts", YEAR=1993, KEYWORDS="routing; MDP; virtual path; Markov Decision Process; VP", ABSTRACT="The virtual path (VP) concept has been proposed to simplify traffic control and resource management in future BISDN. An advantage of this concept is that call setup processing can be significantly reduced when resources are reserved on VP's. However, this advantage can be offset by the decrease in the statistical multiplexing gains of networks supporting VP's. The focus of this paper is on how to improve bandwidth efficiency through adaptive routing when capacity is reserved on VP's. We first examine two VP capacity reservation strategies. We then design and evaluate computationally feasible Markov decision process-based routing algorithms and show that the network call blocking probability can be significantly reduced by MDP routing.", URL="ftp://gaia.cs.umass.edu/pub/Hwang93:MDP.ps.Z", } @ARTICLE{Whit93:CDMA, AUTHOR="Joseph F. White", TITLE="What is {CDMA?}", JOURNAL="Applied Microwave and Wireless", PUBLISHER="J. F. White Publications", PAGES="5-8", MONTH="Fall", YEAR=1993, KEYWORDS="CDMA; cellular phone; modulation; spread spectrum", } @TECHREPORT{Claf93:Measurement, AUTHOR="Kimberly Claffy and George C. Polyzos and H. Braun", TITLE="Measurement Considerations for Assessing Unidirectional Latencies", INSTITUTION="Computer Systems Laboratory, UCSD", ADDRESS="San Diego, California", NUMBER="UCSD Report CS92-252", NOTE="also Journal of Internetworking, 4(3), September 1993", YEAR=1993, ABSTRACT="This paper presents a study of single direction latencies to selected destinations of the Internet utilizing a variety of paths. The objective is to demonstrate that round-trip latencies are an insufficient and sometimes misleading method to determine unidirectional delays. This claim has significant implications for high-speed, multi-application, wide-area, traffic aggregating networking environments which often require predictability of precise delay.", ANNOTE="Assuming that a round trip path is symetric and dividing a RTT in half to find the latency in either the outgoing or incoming tranmission path is highly inaccurate as there are both static and dynamic reasons for the paths traversed by packets between hosts to be asymetric. The static reason is that the route may be permanently configured to be asymetric because of network harware limitations (for example simplex links). The dynamic reasons for asymetrically datagram paths are resource contentions between other components in the network. Digitial continuous media networked applications require good predicatability of unidirectional network delays, which simply halving the RTT cannot give.", URL="ftp://ftp.sdsc.edu/pub/sdsc/anr/papers/single.ps.Z", } @ARTICLE{Demp93:Research, AUTHOR="Lorcan Dempsey", TITLE="Research networks and academic information services: towards an academic information infrastructure: Part 1", JOURNAL="Journal of Information Networking", VOLUME=1, NUMBER=1, YEAR=1993, ANNOTE="SuperJANET will have backbone rings operating at 622Mbps. with 6 pilot sites operating at 140Mbps and a larger number of smaller sites operating at 10Mbps.", } @INPROCEEDINGS{Labo9303:Logical, AUTHOR="Jean-Francois Labourdette and A. S. Acampora", TITLE="Logical Clustering for the Optimization and Analysis of a Rearrangeable Distributed {ATM} Switch", BOOKTITLE=infocom, ADDRESS="San Francisco, California", MONTH="March/April", YEAR=1993, ABSTRACT="This paper describes a practical approach for the analysis and operation of medium to large size distributed ATM switches based on an original optical reconfigurable architecture. The optical nature of the switch provides enormous bandwidth with input/output port rates of the order of Gbits/sec., and the distributed approach is highly modular, creating an easily growable switch. Furthermore, the interconnection pattern between input and output ports is dynamically reconfigurable; the switch can thus exploit nonuniformities in the offered traffic and optimize its connectivity accordingly. For a $24 \times 24$ switch, this leads to improved performance that match those of a fully connected output buffered ideal centralized switch. In medium to large reconfigurable switches, clustering techniques are required for optimizing the connection diagram among modules and routing traffic through the switch. While clustering will save network resources, it will also result in some performance degradation due to non optimal connectivity and routing. However, simulation results for a switch of size $81 \times 81$ seem to indicate that the traffic-handling capability scales well with the size of the switch when non-uniformity prevails in the offered traffic, a realistic assumption. Furthermore, clustering techniques would permit the optimization and operation of much larger reconfigurable distributed ATM switches.", } @INCOLLECTION{Cerf93:Core, AUTHOR="V. G. Cerf", TITLE="Core Protocols", BOOKTITLE="Internet System Handbook", EDITOR="Daniel C. Lynch and Marshall T. Rose", PUBLISHER="Addison Wesley", ADDRESS="Reading, Massachusetts", YEAR=1993, } @TECHREPORT{Claf93:Application, AUTHOR="Kimberly Claffy and H. Braun and George C. Polyzos", TITLE="Application of Sampling Methodologies to Wide-Area Network Traffic Characterization", INSTITUTION="UCSD", NUMBER="Technical Report CS9", YEAR=1993, } @TECHREPORT{Cox93:Market, AUTHOR="B. J. Cox", TITLE="Market Processes as a New Foundation for Software Engineering", INSTITUTION="George Mason University", YEAR=1993, } @MISC{Curr93:Personal, AUTHOR="John Curran", TITLE="Personal communication", YEAR=1993, } @TECHREPORT{Goff93:Internet, AUTHOR="William Goffe", TITLE="{Internet} Resources for Economists", INSTITUTION="University of Southern Mississippi", YEAR=1993, URL="gopher://niord.shsu.edu/11gopher\_root\%3a\%5b\_DATA.ECONOMICS\%5d", } @INPROCEEDINGS{Jami93:Admission, AUTHOR="S. Jamin and L. Zhang and D. D. Clark and S. Shenker", TITLE="An Admission Control Algorithm for Predictive Real-Time Service", BOOKTITLE="3rd International Workshop on Networking and Operating System Support for Digital Audio and Video", PUBLISHER="Springer-Verlag, New York", YEAR=1993, } @TECHREPORT{MacK93:Local, AUTHOR="Jeffrey MacKie-Mason and Donna Lawson", TITLE="Local Telephone Demand with Optional and Nonlinear Prices", INSTITUTION="University of Michigan", YEAR=1993, } @TECHREPORT{MacK93:Economics, AUTHOR="Jeffrey MacKie-Mason and Hal R. Varian", TITLE="Some Economics of the {Internet}", INSTITUTION="University of Michigan", YEAR=1993, } @TECHREPORT{Kosc93:Efficient, AUTHOR="M. Koschat and P. Srinagesh and L. Uhler", TITLE="Efficient price and capacity choices under uncertain demand --an empirical analysis", INSTITUTION="Yale School of Management and Bellcore", YEAR=1993, } @INCOLLECTION{Lync93:Historical, AUTHOR="Dan Lynch", TITLE="Historical Evolution", BOOKTITLE="Internet System Handbook", EDITOR="Daniel C. Lynch and Marshall T. Rose", PUBLISHER="Addison Wesley", ADDRESS="Reading, Massachusetts", YEAR=1993, } @BOOK{Nagu93:Economics, AUTHOR="Anna Nagurney", TITLE="Network Economics", PUBLISHER="Kluwer Academic Publishers", ADDRESS="Boston", YEAR=1993, } @INCOLLECTION{Econ93:Compatibility, AUTHOR="Nicholas Economides", TITLE="Compatibility and the Creation of Shared Networks", BOOKTITLE="Electronic Service Networks", EDITOR="Margaret E. Guerin-Calvert and Steven S. Wildman", PUBLISHER="Praeger", ADDRESS="New York", PAGES="39-55", YEAR=1993, } @UNPUBLISHED{Roma93:Dynamics, AUTHOR="Allyn Romanow and S. Floyd", TITLE="Dynamics of {TCP} Traffic over {ATM} Networks", INSTITUTION="Sun Microsystems", ADDRESS="Mountain View, California", NOTE="see also 6th IEEE LAN/MAN Workshop", YEAR=1993, REFERENCES=41, KEYWORDS="TCP; ATM; congestion; packet dropping; congestion avoidance; packet loss; goodput", ABSTRACT="We investigate the performance of TCP connections over ATM networks with no ATM-level congestion control, and compare it to the performance of TCP over packet-based networks. For simulations with congestion, the effective throughput of TCP over ATM can be quite low when cells are dropped at the congested ATM switch. The low throughput is due to wasted bandwidth as the congested link transmits cells from 'corrupted' packets, packets containing cells that have already been dropped at the switch. This effect can be corrected, and high throughput obtained, if the switch monitors the size of the output buffer queue and drops whole packets when the queue reaches a threshold level; we call this strategy early packet discard. We also discuss general issues of congestion avoidance for best-effort ATM traffic.", URL="ftp://playground.sun.com/pub/tcp\_atm/tcpatm\_extended.00.ps.Z", } @ARTICLE{Schi93:Verifying, AUTHOR="I. Schieferdecker", TITLE="Verifying Performance Oriented Properties of {TIS} Specifications", JOURNAL="Journal of Mathematical Modelling and Simulation in Systems Analysis (SAMS)", VOLUME=13, PAGES="139-162", YEAR=1993, KEYWORDS="timed interacting systems; quality of service; real-time verification; send-and-wait protocol", ABSTRACT="With the increasing bandwidth of communication networks and increasing variety of applications the impact of {\em Quality of Service} (QoS) parameters becomes more and more important. In this paper we use a performance--oriented process algebraic calculus called {\em Timed Interacting Systems} (TIS) for the verification of bounded performance properties, a special kind of QoS properties. Requirements on bounded performance properties can equivalently be described as sets of timing deadlines for that system. A timing deadline demands that time distances between certain events are always within a given time interval, i.e. that events do not occur too early and not too late. We decided to express timing deadlines in a discrete--time linear temporal logic. For the verification, whether a given TIS specification satisfies some timing deadlines, we represent its dynamical behaviour by a finite state machine. The TIS specification has to fulfill three conditions to do this: time progress, bounded monotonicity, and lack of recursive process instantiations under parallel union. We annotate already in the system specification actions by monitoring events defining the time points, from which the time distances for the timing deadlines have to be evaluated. Finally we give decision procedures to show, that the underlying finite state machine, i.e. the TIS specification itself, satisfies the required timing deadlines. In order to demonstrate the application of TIS for the verification of bounded performance properties we investigate a classical Send--and--Wait protocol.", URL="ftp://ftp.fokus.gmd.de/pub/step/papers/Schi.XII93.Verifying.ps.gz", } @ARTICLE{Woli93:Unified, AUTHOR="Adam M Wolisz", TITLE="A Unified Approach to Formal Specification of Communication Protocols and Analysis of their Performance", JOURNAL="Journal of Mathematical Modelling and Simulation in Systems Analysis (SAMS)", VOLUME=13, PAGES="59-88", YEAR=1993, KEYWORDS="timed interacting systems; communication protocols; performance analysis; specification of communication protocols", ABSTRACT="Different description techniques have been traditionally used either for the formal description of communication protocols or for the analysis of their performance. Recently, however, the idea to develop an approach which could be used for both these purposes seems to gain acceptance. Timed Interacting Systems (TIS), a proposal for extending the classical framework of process algebras is presented. Means for expressing the duration of events and time-oriented conditions for the occurrence of events are provided. In addition quantifying the nondeterminism in system behavior is achieved through quantifying the internal choice operation and through introducing logical resources with a priority access scheme. Although in general the model of full parallelism is supported, TIS offers also means for controlled reduction of the potential level of parallelism, making it possible to reflect constraints introduced by an execution environment. Finally formulating performance oriented statements and performance studies are enhanced by the concept of a monitoring environment, supporting the observation of events and intervals between events.", } @TECHREPORT{Schi9307:Verification, AUTHOR="I. Schieferdecker", TITLE="On the Verification of deterministic Bounds for Quality-of-Service Characteristics of Communication Protocols", TYPE="Forschungsprojekte des Graduiertenkollegs-Kommunikationbasierte Systeme", INSTITUTION="TU Berlin, Fachbereich Informatik", ADDRESS="Berlin, Germany", NUMBER=19, YEAR=1993, URL="ftp://ftp.fokus.gmd.de/pub/step/papers/Schi.VII93.Verification.ps.gz", } @INPROCEEDINGS{Gute93:Sharing, AUTHOR="Thomas Gutekunst and B. Plattner", TITLE="Sharing Multimedia Applications Among Heterogeneous Platforms", BOOKTITLE="2nd International Conference on Broadband Islands", ADDRESS="Athens, Greece", YEAR=1993, KEYWORDS="JVTOS; multimedia", ABSTRACT="JVTOS (Joint Viewing and Tele-Operation Service), which is being implemented within CEC/RACE II project CIO (R2060), is a new teleservice for high-speed networks allowing physically distributed users to work in a collaborative fashion. Core functionality of JVTOS is provided by a so-called application sharing service which allows multimedia applications to be displayed and interacted with on multiple users' workstations simultaneously. This paper discusses concept and design issues for the JVTOS application sharing service and its implementation on different platforms. The currently supported platforms are Sparc (X Window System), and Macintosh (QuickDraw). The primary aims of the service are to support heterogeneous platforms and to integrate continuous media.", URL="ftp://komsys.ethz.ch/pub/papers/Gutekunst93b.ps", } @INPROCEEDINGS{Cock93:Making, AUTHOR="A. Cockburn and Saul Greenberg", TITLE="Making contact: getting the group communicating with groupware", BOOKTITLE="COSC", ORGANIZATION="ACM", YEAR=1993, REFERENCES=35, KEYWORDS="contact facilitation; casual interaction; coordination; computer supported cooperative work; groupware.", ABSTRACT="While groupware is readily available, people on wide area networks --- such as the Internet --- have considerable trouble contacting each other and setting up groupware connections. To pinpoint why this occurs, this paper identifies human factors critical to getting a group communicating through groupware. It addresses how people find suitable partners, and how people choose appropriate communication mediums. These factors are discussed in detail, and form a design foundation for systems that promote social presence and that integrate communication. Existing systems are critically reviewed and shown to be inadequate for general use over a wide area net, for they either do not meet some basic design criteria, or they require a very high technological entry level that is beyond the reach of most computer users. As an alternative, the paper presents the design considerations behind TELEFREEK , a flexible, extensible, and customizable platform for collaboration. Drawing on resources freely available to the Internet community, TELEFREEK assists people making contact with others, and integrates access to common communication facilities.", URL="ftp://ftp.cpsc.ucalgary.ca/pub/projects/grouplab/papers/making-contact.COCS93.ps.Z", } @INPROCEEDINGS{Rose93:Building, AUTHOR="Mark Roseman and Saul Greenberg", TITLE="Building flexible groupware through open protocols", BOOKTITLE="COSC", ORGANIZATION="ACM", YEAR=1993, REFERENCES=23, KEYWORDS="computer supported cooperative work; CSCW; conference control; personalizable groupware; expandability; open protocols; implementation technique", ABSTRACT="This paper presents a technical approach to building flexible groupware applications. Flexibility provides the promise of personalizable groupware , allowing different groups to work with the system in diverse ways which best suit the group's own needs. An implementation technique called open protocols is described, which is a variation of client/server architectures. Open protocols facilitate the addition of group-specific modules long after the system has been created. Three examples illustrating the use of open protocols are presented: floor control, conference registration, and brainstorming. Finally, a number of issues facing the groupware developer using open protocols are addressed, along with strategies that can help in dealing with these issues.", URL="ftp://ftp.cpsc.ucalgary.ca/pub/projects/grouplab/papers/open-protocols.COCS93.ps.Z", } @BOOK{Stall93:SNMP, AUTHOR="W. Stallings", TITLE="{SNMP,} {SNMPv2,} and {CMIP:} the practical guide to network management", ISBN="0-201-63331-0", PUBLISHER="Addison-Wesley", ADDRESS="Reading, Massachusetts", YEAR=1993, KEYWORDS="network management; SNMP; CMIP", } @BOOK{Kess93:ISDN, AUTHOR="G. Kessler", TITLE="{ISDN:} concepts, facilities, and services", EDITION="2nd", ISBN="0-07-034247-4", PUBLISHER="McGraw-Hill", ADDRESS="New York, NY", YEAR=1993, KEYWORDS="ISDN; BISDN; X.25; ATM; signaling; LAPB; LAPD", } @INPROCEEDINGS{Chri93:Writing, AUTHOR="Wayne Christopher", TITLE="Writing Object-oriented Tcl-based Systems using Objectify", BOOKTITLE="Tcl Workshop Proceedings", INSTITUTION="University of California, Berkeley", ADDRESS="Berkeley, California", YEAR=1993, REFERENCES=4, KEYWORDS="Tcl; Tk; C++; OOP", ABSTRACT="This paper describes Objectify, a tool that facilitates the integration of C++ classes into Tcl-based systems. It uses an object model similar to that of Tk, and makes it possible for the programmer to annotate his classes with information about which member functions should be exported as Tcl ``methods'', and which data members should be accessible from Tcl via ``configure'' commands. Objectify automatically generates glue code to interface between Tcl and C++, and also creates documentation files. This approach is fairly light weight and straightforward, and automates the tedious and error prone task of maintaining Tcl command interfaces to lower-level object implementations.", URL="ftp://harbor.ecn.purdue.edu/pub/tcl/workshop/tcl-proceedings.tar.Z", } @ARTICLE{Pete93:Multiplexers, AUTHOR="J. Petersen and T. Gillen", TITLE="Multiplexers Sum {D/D/1} with service in cyclic order and service in the order of arrival", JOURNAL=ett, VOLUME=4, NUMBER=4, PAGES="67-75", YEAR=1993, REFERENCES=12, KEYWORDS="ATM; multiplexer; DI/D/1; waiting time", ABSTRACT="The waiting times of cells in a mux and the output process is analyzed in closed form.", } @ARTICLE{Hill93:CM, AUTHOR="W. D. Hillis and L. W. Tucker", TITLE="The {CM-5} connection machine: a scalable supercomputer", JOURNAL=cacm, VOLUME=36, NUMBER=11, PAGES="31-40", YEAR=1993, REFERENCES=24, KEYWORDS="Computer architecture; parallel computing; supercomputer; MIMD", ABSTRACT="The architecture of the CM-5 is described. Based on SPARC processors and two high-speed interconnection systems, up to 16384 PEs can be interconnected. This gives up to 2048 GFLOPS peak.", } @ARTICLE{Chan93:Telecommunications, AUTHOR="Shih-Fu Chang and Bezalel X Gavish", TITLE="Telecommunications network topological design and capacity expansion: formulations and algorithms", JOURNAL=ts, PUBLISHER="J. C. Balzer", ADDRESS="Basel", VOLUME=1, PAGES="99-131", YEAR=1993, REFERENCES=15, KEYWORDS="Communication network; network topology; network design; network capacity; heuristics", ABSTRACT="This paper formulates the network topology and capacity expansion problem in a teleco. network and suggests a family of heuristics and a dual-based lower bounding procedure for solving ist.", } @ARTICLE{Diou93:Dimensioning, AUTHOR="O. Dioume and F. Soumis", TITLE="Dimensioning Telecommunications networks with budget constraints", JOURNAL=ts, PUBLISHER="J.C. Balzer", ADDRESS="Basel", VOLUME=1, PAGES="49-177", YEAR=1993, REFERENCES=10, KEYWORDS="Communication network; network budget; network planning", ABSTRACT="We present a theroretical study of a modified version of the dimensioning problem for telephone networks where the network does not contain any existing capacity and where the available capital is severely limited, a situation which can occur in private networks as well as in the public networks of developingcountries.", } @ARTICLE{Fisc93:Methodology, AUTHOR="M. J. Fischer and D. Swinsky and D. Garland and L. Stanfel", TITLE="A Methodology for designing large private line transmission networks with multiple facilities", JOURNAL=ts, PUBLISHER="J.C. Balzer", ADDRESS="Basel", VOLUME=1, PAGES="243-261", YEAR=1993, REFERENCES=18, KEYWORDS="Communication network; network topology; network design; network capacity", ANNOTE="Deregulation and the recent advances in telecomm. technologies have significantly reduced the cost of transmission facilities. This reduction has made it attractive for corporation and government agenties to develop their own private line transmission networks. In this paper we present a methology we have used in designing these private line transmission networks.", } @ARTICLE{Gouv93:Comparison, AUTHOR="L. Gouveia", TITLE="A Comparison of directed formulations for the capacitated minimal spanning tree problem algorithms", JOURNAL=ts, PUBLISHER="J.C. Balzer", ADDRESS="Basel", VOLUME=1, PAGES="51-76", YEAR=1993, REFERENCES=4, KEYWORDS="Communication network; network topology; network design; network capacity", ANNOTE="In this paper, we compare linear integer programming directed formulations for the capacitated minimal spanning tree (CMST) problem. This problem is directly related to network design and consists of finding the cheapest way to link a set of terminals at different locations with a central node.", } @ARTICLE{Hanl93:Information, AUTHOR="S. V. Hanly and P. Whiting", TITLE="Information-theoretic capacity of multi-receiver networks", JOURNAL=ts, PUBLISHER="J.C. Baltzer", ADDRESS="Basel", VOLUME=1, PAGES="1-42", YEAR=1993, REFERENCES=14, KEYWORDS="Information theory; mobile radio; spread spectrum", ABSTRACT="We consider multi-receiver networks with diversity reception from an information-theoretic point of view. In particular, we find their capacity and investigate how the frequency spectrum should be allocated to users.", } @ARTICLE{Hant93:Designing, AUTHOR="S. L. Hantler and D. Milch", TITLE="Designing reliable networks with short paths", JOURNAL=ts, PUBLISHER="J.C. Balzer", ADDRESS="Basel", VOLUME=1, PAGES="43-50", YEAR=1993, REFERENCES=4, KEYWORDS="Communication network; network topology; network design; network capacity; shortest path", ABSTRACT="We present a technique for constructing highly reliable networksin which locations consist of pairs of redundant vertices and inwhich every pair of locations has a pair of short disjoint pathsconnecting them.", } @ARTICLE{Hubn93:Finite, AUTHOR="F. H bner and P. Tran-Gia", TITLE="Performance of a Finite Capacity Asynchronous Multiplexer with modulated input", JOURNAL=ts, PUBLISHER="J.C. Baltzer", ADDRESS="Basel", VOLUME=1, PAGES="263-278", YEAR=1993, REFERENCES=15, KEYWORDS="ATM; multiplexer; performance evaluation", ABSTRACT="In this paper, a finite capacity multiplexer in ATM environmentswith modulated deterministic input sources is investigated. We consider three cases of input traffic. The main performance measures in this study are the cell and call blocking probabilities.", } @ARTICLE{Lee93:Asynchronous, AUTHOR="G. Lee and Chong Kwan Un", TITLE="Design and Performance Analysis of an Asynchronous Non-blocking switch with window policy", JOURNAL=ts, PUBLISHER="J.C. Baltzer AG", ADDRESS="Basel", VOLUME=1, PAGES="77-97", YEAR=1993, REFERENCES=8, KEYWORDS="Performance evaluation; packet switching", ABSTRACT="We propose an architecture of an asynchronous non-blocking switch. The asynchronous switch proposed in this paper with its implementation of a window scheme can increase its maximum throughput up to 1 in the case of a minimum changeover time and large packet size. We also investigate the delay characteristicsof the asynchronous switch.", } @ARTICLE{Park93:Delay, AUTHOR="Jongseung Park and K. Kang", TITLE="Delay Analysis for Multidimensional Queueing Processes in {CSMA/CD} Local Area Networks", JOURNAL=ts, PUBLISHER="J.C. Baltzer AG", ADDRESS="Basel", VOLUME=1, PAGES="217-242", YEAR=1993, REFERENCES=28, KEYWORDS="CSMA/CD; LAN; semi Markov process", ABSTRACT="A CSMA/CD local area network consists of a single server (the channel) and multiple interacting queues of messages. The message queueing process is modeled as a multidimensional semi-Markov chain. An effective approximation method to compute the meanpacket delay in equilibrium is developed. We also develop a simulation model to validate approximation results.", } @ARTICLE{Ryde93:Optimal, AUTHOR="T. Ryden", TITLE="Optimal Load Control of {SPC} Switches: Waiting Time Minimization", JOURNAL=ts, PUBLISHER="J.C. Baltzer AG", ADDRESS="Basel", VOLUME=1, PAGES="179-194", YEAR=1993, REFERENCES=18, KEYWORDS="SPC; semi Markov process; performance evaluation", ABSTRACT="We address the problem of load control of a switching node with a centralized processor architecture. The problem of finding regulators that minimize the mean dialtone delay is formulated and solved using a semi-Markov decision process model. We also study the transient properties of the optimal regulators, their behaviour for an MMPP arrival process, and discuss their robustness with respect to various model assumptions.", } @ARTICLE{Sans93:Centralized, AUTHOR="B. Sanso and F. Soumis and M. Gendreau", TITLE="Centralized and Decentralized Stochastic Routing Models in telecommunication networks", JOURNAL=ts, PUBLISHER="J.C. Baltzer AG", ADDRESS="Basel", VOLUME=1, PAGES="133-148", YEAR=1993, REFERENCES=30, KEYWORDS="Stochastic routing; network planning; decentralization; circuit switching", ABSTRACT="With the advent of non-hierarchical routing in circuit-switched telecommunication networks, on-line policies have been developedwith the objective of optimizing some measure of gain or performance. These policies are decentralized. We present a decentralized routing model to be used in network planning. We compare it theoretically and empirically with a centralized multicommodity flow model previously presented.", } @ARTICLE{Taki93:Batch, AUTHOR="T. Takine and Toshiharu Hasegawa", TITLE="A batch {SPP/G/1} queue with multiple vacations and exhaustive service discipline", JOURNAL=ts, PUBLISHER="J.C. Baltzer AG", ADDRESS="Basel", VOLUME=1, PAGES="195-215", YEAR=1993, REFERENCES=15, KEYWORDS="Batch arrival; single server; Poisson process; server vacation", ABSTRACT="This paper considers a batch-arrival single server queueing system with multiple vacations and exhaustive service discipline. Customers arrive to the system in accordance with a batch switched Poisson process. We analyze the stationary queue length distribution and derive various formulas for queue lengths and waiting times.", } @ARTICLE{Zhan93:Modified, AUTHOR="Z. Zhang and A. S. Acampora", TITLE="Performance of a Modified Polling Strategy for Broadband Wireless {LANs} in a Harsh Fading Environment", JOURNAL=ts, PUBLISHER="J. C. Baltzer AG", ADDRESS="Basel", VOLUME=1, PAGES="279-294", YEAR=1993, REFERENCES=13, KEYWORDS="Communication network; radio channel; polling; LAN", ABSTRACT="In this paper, we consider a radio-based communication network in which a single, high speed radio channel is shared by some plurality of small portable ``notebooks''. In this harsh indoor fading environment, a base-station approach is used. We propose a modified polling scheme for the indoor radio LAN channel. The efficiency of such a scheme is studied and the expected turn-around delay as a function of network load is approximately obtained", } @ARTICLE{Brun93:Buffer, AUTHOR="H. Bruneel and Yongqiang Xiong", TITLE="Buffer contents and delay for statistical multiplexers with fixed-length packet-train arrivals", JOURNAL=pe, ADDRESS="Amsterdam", VOLUME=17, NUMBER=1, PAGES="31-42", YEAR=1993, REFERENCES=8, KEYWORDS="Statistical multiplexer; discrete time queue; generating function; packet delay", ABSTRACT={A slotted statistical multiplexer is considered, to which messages containing a fixed number of fixed-length packets each, arrive at the rate of one packet per slot. The multiplexer buffer cantherefore be studied using a discrete-time queueing model with correlated arrivals, referred to as "train arrivals" in the paper. This train arrival model is completely analyzed by means of agenerating-functions approach}, } @ARTICLE{Cowe93:Concurrence, AUTHOR="L. Cowen and R. Mathar", TITLE="Concurrence probabilities for a locally slotted packet radio network by combinatorial methods", JOURNAL=pe, ADDRESS="Amsterdam", VOLUME=17, NUMBER=1, PAGES="43-51", YEAR=1993, REFERENCES=10, KEYWORDS="Slotted ALOHA; throughput; packet radio", ABSTRACT="Locally time-synchronized slot systems for broadcast communication in packet radio networks are considered. To evaluate the performance of such systems, we investigate in this case the probability of colliding slots. In an appropriate probabilistic model, the throughput of the channel is obtained, and this leads to a complete closed form representation of the system performance.", } @ARTICLE{Lee93:Transient, AUTHOR="D. Lee and Song Li", TITLE="Transient analysis of a switched Poisson arrival queue under overload control", JOURNAL=pe, ADDRESS="Amsterdam", VOLUME=17, NUMBER=1, PAGES="13-29", YEAR=1993, REFERENCES=27, KEYWORDS="Transient analysis; SPP; switched poisson process; integrated service; overload control; packet switching; finite state Markov chain", ABSTRACT="This paper studies the transient behaviour of a switched Poissonarrival queue under overload control. The queue has infinite buffer capacity with an exponential server. Our study provides important information to overload control design on packet switching networks. We also explore the transient queueing behaviour as affected by time stochastic properties of the underlyingtwo-state Markov chain for the arrival process.", } @ARTICLE{Nels93:Approximation, AUTHOR="Randolph D. Nelson and T. K. Philips", TITLE="An approximation for the mean response time for shortest queue routing with general interarrival and service times", JOURNAL=pe, ADDRESS="Amsterdam", VOLUME=17, NUMBER=2, PAGES="123-139", YEAR=1993, REFERENCES=21, KEYWORDS="Load balancing; interarrival time; Poisson process", ABSTRACT="In this paper we derive an approximation for the mean response time of a multiple queue system in which shortest queue routing is used. We assume there are $K$ identical queues with infinite capacity. Interarrival and service times are generally distributed, and an arriving job is routed to a queue of minimal length.", } @ARTICLE{Sztr93:Modelling, AUTHOR="K. Sztrik", TITLE="Modelling of a multiprocessor system in a randomly changing environment", JOURNAL=pe, ADDRESS="Amsterdam", VOLUME=17, NUMBER=1, PAGES="1-11", YEAR=1993, REFERENCES=19, KEYWORDS="Multiprocessor system; performance measure; utilization; exponential distributed transition time", ABSTRACT="A queueing-theoretic approach is developed to analyse the performance of homogeneous multiprocessor computer systems evolving ina random environment. Moreover, exact and approximate validation results are presented to illustrate the credibility of the proposed method.", } @ARTICLE{Yama93:Cell, AUTHOR="Hiroshi Yamada", TITLE="Cell/packet loss behavior in a statistical multiplexer with bursty input", JOURNAL=pe, ADDRESS="Amsterdam", VOLUME=17, NUMBER=2, PAGES="77-99", YEAR=1993, REFERENCES=30, KEYWORDS="Statistical multiplexer; bursty traffic; ATM; loss probability; performance measure; performance evaluation", ABSTRACT="Due to high correlation between consecutive cells or packets in ATM networks or high-speed packet networks, the average cell/packet loss probability does not well characterize the real cell/packet loss behavior. Other performance measures are therefore needed, and this paper considers the lossy period, the non-lossyperiod, the inter-overflow times, and the moment of the number of lost or arriving cells or packets during the lossy period.", } @ARTICLE{Lore93:Vergleich, AUTHOR="R. W. Lorenz", TITLE="Vergleich der digitalen Mobilfunksysteme in Europa {(GSM)} und Japan {(JDC)} unter besonderer Berücksichtigung der Wirtschaftlichkeitsaspekte", JOURNAL=fernmelde, PUBLISHER="Verlag für Wissenschaft und Leben, Heidecker", ADDRESS="Erlangen", VOLUME=17, NUMBER="1-2", PAGES="1-72", YEAR=1993, LANGUAGE="German", REFERENCES=104, KEYWORDS="Mobile radio; comparison; digital communication", ABSTRACT="Die technischen Unterschiede der Funkschnittstellen des GSM- und des JDC-Systems ergeben für letzteres eien größere Teilnehmerkapazität. Allerdings macht das JDC-System teilweise die Anwendung von Komponenten notwendig, die ein höheres Entwicklungsniveau haben müssen als die im GSM-System. Es werden die Vor- und Nachteile der technischen Lösungen unter Berücksichtigung der Wirtschaftlichkeit verglichen.", } @ARTICLE{Pere93:Distribution, AUTHOR="R. Perera and F. Schreiber", TITLE="Distribution function and local correlation of the delay time in queueing system {M/G/1-SRPT}", JOURNAL=aeu, PUBLISHER="Hirzel-Verlag", VOLUME=47, NUMBER=2, PAGES="108-110", YEAR=1993, REFERENCES=10, KEYWORDS="Queueing system; M/G/1; simulation; SRPT; shortest remaining processing time", ABSTRACT="The statationary distribution function of delay time has been obtained by simulation. Shows the superiority of SRPT in comparison to classical strategies FIFO and LIFO. A further essential advantage: the practically zero local correlation coefficient of delay time.", } @ARTICLE{Uhl93:Warteschlangenanalyse, AUTHOR="T. Uhl", TITLE="Warteschlangenanalyse von Datennetzen mit begrenzten Ressourcen: Ein Weg zur Leistungsanalyse der Routingtechniken", JOURNAL=aeu, PUBLISHER="Hirzel-Verlag", VOLUME=47, NUMBER=2, PAGES="91-97", YEAR=1993, LANGUAGE="German", REFERENCES=13, KEYWORDS="Queueing network; routing; packet switching; performance evaluation", ABSTRACT="This study provides the development of an analytical method for a performance analysis of loop free routing techniques in data networks with limited resources.", } @ARTICLE{Sczi93:Aggregation, AUTHOR="M. Sczittnick and T. Uhl", TITLE="Aggregation of systems with {HDLC} protocol", JOURNAL=ett, ADDRESS="Milan", VOLUME=4, NUMBER=1, PAGES="107-112", YEAR=1993, REFERENCES=16, KEYWORDS="Queueing network; protocol performance; HDLC; aggregation", ABSTRACT="In this paper an aggregated mathematical model for the performance analysis of a node with HDLC protocol will be presented that is still accurate concerning mean delay and losses.", } @ARTICLE{Lee93:Virtual, AUTHOR="K.-H. Lee", TITLE="A virtual bus architecture for dynamic parallel processing", JOURNAL=ieeepds, VOLUME=4, NUMBER=2, PAGES="121-130", YEAR=1993, REFERENCES=40, KEYWORDS="Parallel computing; interconnection network; bus; arbitration; performance evaluation", ABSTRACT="We propose a large scale interconnection network architecture called a virtual bus. The virtual bus can scale to terabits per second end-to-end communication bandwidth with low queueing delay for nonuniform traffic.", } @ARTICLE{Chla93:HIPPI, AUTHOR="I. Chlamtac and A. Ganz and M. G. Kienzle", TITLE="An {HIPPI} interconnection system", JOURNAL=ieeetc, VOLUME="C-42", NUMBER=2, PAGES="138-150", YEAR=1993, REFERENCES=14, KEYWORDS="Parallel computing; interconnection network; performance evaluation", ANNOTE="This paper deals with the proposed HIPPI standard, a connection-oriented standard for very high-speed systems.", } @MISC{Addi93:Queues, AUTHOR="R. G. Addie and Moshe Zukerman", TITLE="Queues with Total Recall", PAGES=15, YEAR=1993, REFERENCES=13, KEYWORDS="Semi Markov Process; Autoregressive Process; G/G/1; Queueing System; Single Server; Statistical Multiplexing; Performance Evaluation; Approximation; Loss Probability; Connection Admission Control", ABSTRACT="Queues in which the past of the input process remains relevant to its future evolution for a long time (queueing systems with 'total recall') can be analysed by means of the Wiener-Hopf factorization technique. Under suitable conditions, the stationary distribution of queued work has a tail of exponentialform and in many cases this tail is a good approximation to the entire distribution. Explicit results are given for the case", } @INPROCEEDINGS{Addi9303:Approximation, AUTHOR="R. G. Addie and Moshe Zukerman", TITLE="An Approximation for Performance Evaluation of Stationary Single server Queues", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=2, PAGES="835-842", MONTH="March/April", YEAR=1993, REFERENCES=25, KEYWORDS="Semi Markov Process; Autoregressive Process; G/G/1; Queueing System; Single Server; Statistical Multiplexing; Performance Evaluation; Approximation; Loss Probability", ABSTRACT="This paper provides a method for approximating the probability distributions of stationary statistics in FIFO single server queues. The method is based on the Wiener-Hopf factorization technique, and is applied to semi-Markov queues where the underlying state space is of unlimited size. A particularly important case is a model for a statistical multiplexer where the net input process forms a stationary ergodic Gaussian", } @ARTICLE{Wach93:Breitband, AUTHOR="G. Wachholz and W. Falkenburg and J. Fingerholz and others", TITLE="Das Breitband-ISDN-Pilotprojekt der Telekom", JOURNAL=fernmelde, VOLUME=47, NUMBER="10/11", PAGES="32+15", YEAR=1993, LANGUAGE="German", REFERENCES=8, KEYWORDS="ATM; BISDN; test", } @INPROCEEDINGS{Agra9303:Local, AUTHOR="Gopal Agrawal and Baio Chen and W. Zhao", TITLE="Local Synchronous Capacity Allocation Schemes for Guaranteeing Message Deadlines with the Timed Token Protocol", BOOKTITLE=infocom, ADDRESS="San Francisco", PAGES="186-193", MONTH="March/April", YEAR=1993, REFERENCES=43, KEYWORDS="FDDI; MAN; allocation; synchronous transmission", ABSTRACT="In this paper, we develop and analyze a class of local synchronous capacity allocation schemes. A local scheme allocates the synchronous capacity to a node without using information about messages on the other nodes while a global allocation scheme uses network wide information in the allocation process. Use of local schemes benefits the run-time management of network in the sense that the entire network can continue its normal operation.", } @INPROCEEDINGS{Anas9303:Service, AUTHOR="G. Anastasi and M. Conti and Enrico Gregori and Luniano Lenzini", TITLE="Service Integration in {CRMA:} A Simulative Analysis", BOOKTITLE=infocom, ADDRESS="San Francisco", PAGES="715-721", MONTH="March/April", YEAR=1993, REFERENCES=13, KEYWORDS="MAN; CRMA; cyclic reservation multiple access; simulation; VBR; variable bitrate coding; service integration", ABSTRACT="This paper reports on a simulative analysis of the integration of VBR video and voice traffics in a CRMA network. The most original aspect of the paper is the characterization of the video sources via a trace of the output of a real codec. In this paper we show that by managing internet voice as VBR video the delay problem of voice traffic is overcome. Furthermore, we analyze the dependency of the QoS parameter values on the video source chara.", } @INPROCEEDINGS{Deng93:Optimal, AUTHOR="S. Deng and U. M. Maydell", TITLE="Optimal control of flexible bandwidth calls in {BISDN}", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=3, PAGES="1315-1319", YEAR=1993, REFERENCES=16, KEYWORDS="BISDN; bandwidth allocation; network management", ABSTRACT="Optimal admission control is an important network management andcontrol problem. Finding an optimal solution, however, is computation prohibitive in broadband networks. Optimal admission control becomes feasible in practical applications only if optimal admission policies exhibit certain policy structures so that thesearch for optimality is constrained. The paper proves the existnce of a threshold structure for networks serving traffic mix", } @INPROCEEDINGS{Bann9303:Multifiber, AUTHOR="J. Bannister and M. Gerla and Milan Kovacevic", TITLE="An All-Optical Multifiber Tree Network", BOOKTITLE=infocom, ADDRESS="San Francisco", PAGES="282-292", MONTH="March/April", YEAR=1993, REFERENCES=21, KEYWORDS="Optical LAN; tree network; MAN; hslan", ABSTRACT="We present a network architecture that uses the multifiber tree topology to provide high-speed datagram and circuit-switching communication services to a large population of stations. Using a combination of space-, wavelength-, and time-division multiplexing, this network architecture provides all-optical transmission media to its stations and can interconnect several thousand stations in a metropolitan region.", } @ARTICLE{Blou93:Search, AUTHOR="P. L'Ecuyer and F. Blouin and R. W. Couture", TITLE="A search for good multiple recursive random number generators", JOURNAL=tomacs, ADDRESS="New York", VOLUME=3, NUMBER=2, PAGES="87-98", YEAR=1993, REFERENCES=24, KEYWORDS="RNG; linear congruential generator; pseudo-random generator; ecursive random generator; spectral test", ABSTRACT="We report the results of an extensive computer search for good multiple recursive generators, in terms of their lattice structure and implementation speed. Thode generators are a little slower than the ususal linear congruential generators, but have much longer periods and much better statistical properties. We provide specific parameter sets for 32-bit, 48-bit, and 64-bit computers.", } @ARTICLE{Brem93:Derivatives, AUTHOR="P. Bremaud and Wei-Bo Gong", TITLE="Derivatives of likelihood ratios and smoothed pertubation analysis for the routing problem", JOURNAL=tomacs, ADDRESS="New York", VOLUME=3, NUMBER=2, PAGES="134-161", YEAR=1993, REFERENCES=22, KEYWORDS="Discrete event simulation; estimation; sensitivity analysis; rgodicity", ABSTRACT="In this article we present estimates for the gradients of the cycle variables with respect to a thinning parameter in the arrival process of G/G/1 queueing systems. Our estimates belong to the category of the likelihood ratio method, smoothed-pertubation analysis, and rare-pertubation analysis. In particular, we obtain unbiased estimates of derivatives of any order in the random-horizon case and the stationary and ergodic estimates.", } @ARTICLE{Raat93:Sequential, AUTHOR="K. Raatikainen", TITLE="A sequential procedure for simultaneous estimation of several means", JOURNAL=tomacs, ADDRESS="New York", VOLUME=3, NUMBER=2, PAGES="108-133", YEAR=1993, REFERENCES=34, KEYWORDS="Estimation; simulation; confidence interval", ABSTRACT="Sequential procedures for controlling the length of a simulationrun are widely used when a single mean is estimated. In many practical situations, however, the analyst is simultaneously interested in several means. We propose a sequential procedure for controlling the length of a simulation run when several means are simultaneously estimated.", } @ARTICLE{Walk93:Mobile, AUTHOR="B. Walke and P. Decker", TITLE="Mobile Datenkommunikation - Eine Übersicht", JOURNAL=itti, PUBLISHER="R. Oldenburg Verlag", VOLUME=35, NUMBER=5, PAGES="12-25", YEAR=1993, LANGUAGE="German", REFERENCES=21, KEYWORDS="Mobile radio; data communication; survey", ABSTRACT="Systems technical parameters, modes of operation and state of introduction of mobile radio networks are described.", } @INPROCEEDINGS{Bern9303:Source, AUTHOR="F. Bernabei and Laura Gratta and M. Listanti and A. Sarghini", TITLE="Analysis of {ON-OFF} Source Shaping for {ATM} Multiplexing", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="1330-1336", MONTH="March/April", YEAR=1993, REFERENCES=8, KEYWORDS="ATM; Traffic Control; Bursty Traffic; statistical multiplexer; controlled input", ABSTRACT="This paper deals with the multiplexing of shaped ON-OFF sources on an ATM link. In particular, a (N)-Level Shaper is analysed. It modulates the source output bit rate on the basis of the buffer occupancy level. A very accurate analytical model describing its behaviour is provided. Numerical results show theeffectiveness of the proposed shaping mechanism.", } @ARTICLE{Kuma93:Scheduling, AUTHOR="Anurag Shorey and R. Krishna Kumar", TITLE="Performance analysis and scheduling of stochastic fork-join jobs in a multicomputer system", JOURNAL=ieeepds, VOLUME=4, NUMBER=10, PAGES="1147-1164", YEAR=1993, KEYWORDS="Performance evaluation; queueing system; fork-join queue; parallel computing; queueing discipline", } @ARTICLE{Bogi93:Low, AUTHOR="K. Bogineni and Krishna M. Sivalingam and P. W. Dowd", TITLE="Low-Complexity Multiple Access Protocols for Wavelength-Division Multiplexed Photonic Networks", JOURNAL=ieeejsac, VOLUME="SAC-11", NUMBER=4, PAGES="590-604", YEAR=1993, REFERENCES=42, KEYWORDS="Star network; protocol; HSLAN; WDM; wavelength division multiplex", ABSTRACT="Media access control protocols for an optically interconnected star-coupled system with preallocated wavelength-division multiple-access channels are introduces and compared. (I-SA, I-TDMA)", } @INPROCEEDINGS{Bohm93:Prioritized, AUTHOR="Stephen Bohm and A. K. Elhakeem and V. K. Murthy", TITLE="Performance of a prioritized nonblocking Switch under Hotspot traffic Conditions with Head of Line Resolution", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=2, PAGES="686-690", YEAR=1993, REFERENCES=7, KEYWORDS="Performance evaluation; HOL priority; nonblocking; priority; ot spot", ABSTRACT="In this paper, we pursue a performance analysis under hotspot traffic conditions on a four level prioritized nonblocking baseband switch for use on-board a switching multibeam satelliteBoth finite input and output buffering as well as speedup are employed to reduce the loss which is critical in a satellite application.", } @ARTICLE{Boli93:SBP, AUTHOR="S. Bolis and E. G. Economou and D. Mouzakis and G. Philokyprou", TITLE="{SBP-Net:} an integrated voice/data token ring {LAN}", JOURNAL=ccr, VOLUME=16, NUMBER=8, PAGES="494-500", YEAR=1993, REFERENCES=14, KEYWORDS="Token ring; LAN; voice communication; voice/data integration", ABSTRACT="This paper presents a modified token ring LAN, called the Scheduling Bypass Periods Network (SBP-Net). SBP-Net uses the ring to implement the IEEE 802.5 standard token ring functions and an auxiliary bus for control purposes. In our proposal idle stations are bypassed, therefore the relevant delay produced in the standard token ring is minimized. This improvement is paricularly significant at light and medium loads.", } @INPROCEEDINGS{Bono9303:One, AUTHOR="A. Bononi and F. Forghieri and P. R. Prucnal", TITLE="Analysis of One-Buffer Deflection Routing in Ultra-Fast Optical Mesh Networks", BOOKTITLE=infocom, ADDRESS="San Francisco", PAGES="303-311", MONTH="March/April", YEAR=1993, REFERENCES=12, KEYWORDS="MAN; shufflenet; routing; analysis", ABSTRACT="The steady state behaviour of regular two-connected multihop networks in homogeneous load under hot-potato and single-buffer deflection routing is analyzed for ultra-fast optical applications. Manhatten Street Network and ShuffleNet are compared in terms of throughput, delay and deflection probability both analytically and by simulation.", } @ARTICLE{Bore93:Enhancements, AUTHOR="A. Borella and D. Broglio and G. Cancellieri and F. Chiaraluce", TITLE="Enhancements of {DQDB} protocol with {ECBWB} mechanism for fair access and multi-priority traffic management", JOURNAL=ccr, VOLUME=16, NUMBER=8, PAGES="511-517", YEAR=1993, REFERENCES=7, KEYWORDS="Man; DQDB; bandwidth allocation; fairness; multiple priority", ABSTRACT="The Distributed Queue Dual Bus (DQDB) protocol with BandWidth Balancing (BWB) mechanism is unable to effectively support multiple priority levels that may be necessary for providing new services to network users. Thus, fair allocation of bandwidth among users is assured only in the case of one level of traffic priority. To overcome this limitation, this paper introduces a novel mechanism called Extended Counting BandWidth Balancing (ECBWB).", } @INPROCEEDINGS{Box9303:Architecture, AUTHOR="Donald F. Box and D. Hong and T. Suda", TITLE="Architecture and Design of Connectionless Data Service for a public {ATM} Network", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=2, PAGES="722-731", MONTH="March/April", YEAR=1993, REFERENCES=17, KEYWORDS="ATM; Connectionless Service; Datagram; Architecture", ABSTRACT="The deployment of a high speed, wide area ATM network introducesa 'backwards compatibility' dilemma. While emerging ATM networksmust be capable of transporting traffic sources such as digitized voice and video, many of the design choices that optimize ATM for these sources make the transport of traditionalconnectionless data traffic difficult. This paper proposes the overlay of a virtual datagram network of connectionless servers", } @ARTICLE{Buss93:Zwei, AUTHOR="H. Busse", TITLE="M-/S-Bus, zwei flotte Geschwister", JOURNAL="Elektronik", VOLUME=5, PAGES="84-87", YEAR=1993, LANGUAGE="German", REFERENCES=1, KEYWORDS="Bus; throughput", ABSTRACT="Prozessoren, die mit großen Bandbreiten und hohen taktraten arbeiten, werden gern als Leistungsmerkmal für Rechner angeführt. Der Rechnerarchitektur, und speziell den internen oder externen Kommunikationskanälen zwischen CPU und Peripherie, sollte jedoch mehr Aufmerksamkeit gewidmet werden.", } @INPROCEEDINGS{Chan93:ATM, AUTHOR="C. J. Chang and Jinghu Chen and Philip Lin", TITLE="Analysis of an {ATM} multiplexer with correlated periodic real-time and independent batch Poisson non-real time traffic", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=2, PAGES="1032-1036", YEAR=1993, REFERENCES=18, KEYWORDS="ATM; multiplexer; batch arrival; Poisson process; periodic input traffic", ABSTRACT="This paper presents an alternate traffic model for an ATM multiplexer providing video, voice, and data services. The traffic model categorizes input traffic to the ATM multiplexer into two types: real-time and nonreal-time, where the real-time traffic is periodic and correlated, while the nonreal-time traffic is batchPoisson and independent. This multiplexer is assumed to be a queueing system with synchronous servers", } @INPROCEEDINGS{Chen93:Finite, AUTHOR="Chi-Ming Chen and Shuo-Hsien Hsiao and D. Meliksetian and K. C. Nwosu", TITLE="Performance Analysis of finite-buffered multistage interconnection networks", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=1, PAGES="53-57", YEAR=1993, REFERENCES=9, KEYWORDS="ATM; Performance evaluation; finite buffer; multistage interconnection network", ABSTRACT="A new model for the performance evaluation of finite buffered multistage interconnection networks (MINs) is proposed. In contrast to previous models, which are either rather inaccurate when input load is high or only applicable to some special cases (...), the proposed model is very accurate for all ranges of input load and allows switching elements and buffer modules to be of any arbitrary size.", } @ARTICLE{Chen93:Distributed, AUTHOR="Wei-Peng Chen and Jiunn-Hwa. Yu", TITLE="Distributed protocol for integrated voice/data token passing ring networks", JOURNAL=comcom, ISBN="0140-3664", VOLUME=16, NUMBER=6, PAGES="338-349", YEAR=1993, REFERENCES=23, KEYWORDS="Distributed processing; performance improvement; synchronous transmission; token ring", } @INPROCEEDINGS{Chen93:Modeling, AUTHOR="Xiaopeng Chen", TITLE="Modeling Connection Admission Control", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=1, PAGES="274-281", MONTH="March/April", YEAR=1993, REFERENCES=9, KEYWORDS="Statistical Multiplexer; Bursty Traffic; MMPP/G/1; Queueing System; Finite Queue; Analysis; Approximation; Loss Probability; IPP; interrupted Poisson Process", ABSTRACT="This paper presents the performance analysis of connection admission control based on the single-class single-queue scheme whose input consists of statistical identical ON-OFF traffic sources, each modeled by an IPP process. The first part of this paper is concerned with the exact analysis of the scheme, known as the MMPP/G/1/K queue in the literature. The computational complexity associated with the exact analysis becomes", } @INPROCEEDINGS{Chen93:Irregular, AUTHOR="Yunfei Chen and Galen Sasaki", TITLE="Irregular Torus Networks: Deadlock Avoidance and Througput Analysis", BOOKTITLE=infocom, ADDRESS="San Francisco", PAGES="312-321", MONTH="March/April", YEAR=1993, REFERENCES=14, KEYWORDS="Man; deadlock; network; throughput analysis", ABSTRACT="In this paper a buffer management is proposed to prevent store-and-forward deadlock for routing in the irregular torus, and the throughput of a typical irregular torus network is analyzed. Store-and-forward deadlock is avoided by a buffer management whose storage requirement and control complexity at a node are both constant with the size of a network.", } @ARTICLE{Chip93:Protocols, AUTHOR="Renu Chipalkatti and Z. Zhang and A. S. Acampora", TITLE="Protocols for Optical Star-Coupler Network Using {WDM:} Performance and Complexity Study", JOURNAL=ieeejsac, VOLUME="SAC-11", NUMBER=4, PAGES="579-589", YEAR=1993, REFERENCES=21, KEYWORDS="Star network; protocol; WDM; wavelength division multiplex", ABSTRACT="In this paper, the performance and complexity issues involved in designing protocols for passive star networks are explored. (DAS, TDM and HTDM)", } @INPROCEEDINGS{Cido93:Protective, AUTHOR="I. Cidon and R. Guérin and Asad Khamisy", TITLE="On Protective Buffer Policies", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="1051-1058", MONTH="March/April", YEAR=1993, REFERENCES=14, KEYWORDS="ATM; Buffer Management; Packet Discarding; Priority; Performance Evaluation; Simulation; Loss Probability", ABSTRACT="We study buffering policies which provide different loss priorities to cells with no change in cell ordering (space priority disciplines). The main issue being investigated is the identification and the evaluation of buffering policies that canguarantee performance, i.e., loss probability, to high priority cells irrespective of the traffic intensity and arrival patternsof low priority packets. Such policies are termed protective", } @INPROCEEDINGS{Cohe93:Simulation, AUTHOR="Danny Cohen and D. P. Heyman", TITLE="A Simulation Study of Video Teleconferencing Traffic in {ATM} networks", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="894-901", MONTH="March/April", YEAR=1993, REFERENCES=4, KEYWORDS="Packet Video; Videoconference; ATM; Traffic Source Model; Markov Chain; Simulation; Loss Probability; Statistical Multiplexing", ABSTRACT="This note presents results of a simulation study of the potential multiplexing gains from using Variable Bit Rate encoding to multiplex video teleconferencing traffic over an ATM network. Simulated traffic from several video teleconferences was fed into a buffer and multiplexed onto a higher speed outputline. The study observed the cell loss resulting from buffer overflow.", } @INPROCEEDINGS{Cohe93:Multiple, AUTHOR="Richard M. Cohen and A. Segall", TITLE="Multiple Logical Token-Rings in a Single High-Speed Ring", BOOKTITLE=infocom, ADDRESS="San Francisco", PAGES="699-706", MONTH="March/April", YEAR=1993, REFERENCES=7, KEYWORDS="Ring network; token ring; token system; MAN; hsLAN; protocol", ABSTRACT="The paper describes a new media access control protocol, called the multiple token protocol, for high-speed ring networks. The purpose of this protocol is to decrease the access delay under a light load of the token ring protocol. The protocol maintains several logical token-rings in the single physical ring. The number of logical rings can be changed dynamically by the ring stations according to the actual ring load.", } @INPROCEEDINGS{Coll93:Path, AUTHOR="M. Collier and T. Curran", TITLE="Path Allocation in a Three-Stage Broadband Switch with intermediate Channel Grouping", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="927-934", MONTH="March/April", YEAR=1993, REFERENCES=12, KEYWORDS="ATM; Switching; Clos; Path Control; Parallel Algorithm; Implementation; VLSI", ABSTRACT="A method for path allocation is described for use with 3-stage ATM switches which feature multiple channels between the switch modules in adjacent stages. The method is suited to hardware implementation using parallelism to achieve a very short execution time. This allows path allocation to be performed anewin each slot.", } @INPROCEEDINGS{Cros93:Renegotiation, AUTHOR="S. Crosby", TITLE="In-Call Renegotiation of Traffic Parameters", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=2, PAGES="638-646", MONTH="March/April", YEAR=1993, REFERENCES=17, KEYWORDS="ATM; Connection Admission Control; Bandwidth Allocation; Dynamic Resource Allocation; Bursty Traffic; Performance Evaluation", ABSTRACT="Static allocation parameterises a source at connection setup only, whereas dynamic allocation requires a source to inform themultiplexer whenever its bandwidth requirements change. We analyse a simple dynamic 'renegotiation' protocol and derive bounds on the performance of an optimal dynamic protocol. Markovian source models parameterised from working applications are used to contrast the behaviour of the three control schemes.", } @ARTICLE{Davi93:Self, AUTHOR="A. C. Davidson and S. K. Chaudhuri", TITLE="A Self-Routing Packet Network Via Optical Processing of the Header", JOURNAL=ett, ADDRESS="Milano", VOLUME=4, NUMBER=2, PAGES="201-211", YEAR=1993, REFERENCES=16, KEYWORDS="HsLAN; optical LAN; optical switching; slotted ring; token ring", ABSTRACT="To allow multiple users access to a high speed optical channel, a new scheme, given the name code-division header recognition (CDHR), is devised that uses both frequency and time-domain bandwidth expansion as well as optical signal processing. An encoded destination address for a data packet is transmitted on a low speed subchannel at a different wavelength from a high speed main channel and optical delay-line correlators are used for routing.", } @INPROCEEDINGS{Deci93:Connectionless, AUTHOR="M. Decina and P. Giacomazzi and Achille Pattavina", TITLE="Connectionless Switching by the Asynchronous Shuffleout Network", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=2, PAGES="701-707", YEAR=1993, REFERENCES=6, KEYWORDS="ATM; connectionless service; self routing", ABSTRACT="A new architecture is proposed in this paper to perform a switching function in a connectionless network environment. The structure is a multistage arrangement of unbuffered switching elements that exploits the packet self-routing technique by attaching a self-routing tag to each packet.", } @INPROCEEDINGS{Deci93:Approximate, AUTHOR="M. Decina and V. Trecordi and G. Zanolini and D. Zucca", TITLE="An Approximate Analysis of Broadcasting in Multichannel Metropolitan Area Networks", BOOKTITLE=infocom, ADDRESS="San Francisco", PAGES="293-302", MONTH="March/April", YEAR=1993, REFERENCES=13, KEYWORDS="Man; broadcast; analysis; multichannel", ABSTRACT="This paper focusses on an open issue in Multyichannel MANs, namely broadcast service. Two different strategies are proposed to deliver of broadcast traffic in Bidirectional Manhattan Street Network (BMSN) and Shuffle Net (SN). The comparions is based on throughput and delivery delay results, as obtained by developing an analytic model. Lights and shadows of the different approaches to the problem of broadcasting, are pointed out.", } @ARTICLE{Deng93:Token, AUTHOR="R. H. Deng and Xianan Zhang and K.-T. Huang", TITLE="Performance of a token-passing system with batch arrivals and its application to file transfers", JOURNAL=ccr, VOLUME=16, NUMBER=7, PAGES="422-431", YEAR=1993, REFERENCES=20, KEYWORDS="Token passing; token system; file transfer; batch arrival; performance analysis", ABSTRACT="This paper investigates the performance of token-passing systems with limited service and Poisson arrivals. For pure Poisson arrivals, the Laplace-SDtieltjes Transform (LST) of an approximate customer/packet waiting time distribution is derived and expressed as a functional equation, from which the approximate mean and variance of waiting time are obtained; for batch Poisson arrivals, an approximate mean of waiting time is derived.", } @INPROCEEDINGS{Ding93:Nonuniform, AUTHOR="J. Ding", TITLE="Nonuniform traffic analysis of multistage interconnection networks with split buffers", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=1, PAGES="58-62", YEAR=1993, REFERENCES=12, KEYWORDS="Multistage interconnection network; traffic analysis; Omega network", ABSTRACT="This paper concerns the analysis of the Split-buffered multistage interconnection networks (SMINs) under the favorite input loadThe favorite load is a typical nonuniform pattern which includesthe uniform load as its special case. A new analytical model of SMINs under the favorite load is developed.", } @ARTICLE{Drak93:Traffic, AUTHOR="E. Drakopolis", TITLE="Performance and Traffic Analysis of a Network-Based Distributed system", JOURNAL=comcom, VOLUME=16, NUMBER=3, PAGES="155-167", YEAR=1993, REFERENCES=11, KEYWORDS="LAN; Traffic Measurement; Performance Evaluation; Queueing Network", ABSTRACT="In this paper we study the performance of a distributed computing system that consists of a number of clients and a server, interconnected by a CSMA/CD LAN. Each client represents a single-user diskless workstation. To characterize the application workload we perform a series of measurement experiments in a controlled environment. The measurements obtained from the experiments are used to parametrize a queueing", } @INPROCEEDINGS{Dzio93:Fair, AUTHOR="L. G. Mason and Z. Dziong and Nicole Tetreault", TITLE="Fair-efficient call admission control policies for broadband networks", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=2, PAGES="976-982", YEAR=1993, REFERENCES=17, KEYWORDS="ATM; connection admission control; broadband; resource utilization", ABSTRACT="A fundamental problem in connection oriented multi-service networks (ATM and STM), is finding the optimal policy for call acceptance. The measure of performance for call class $j$ is $Bj$, the call loss probability. One seeks an admission control policy which efficiently utilizes network resources while at the same time being fair to the various call classes being supported. The theory of co-operative games provides a natural framework.", } @ARTICLE{Eage93:Chores, AUTHOR="D. L. Eager and J. L. Zahorjan", TITLE="Chores: Enhanced Run-Time Support for Shared-Memory Parallel Computing", JOURNAL=tocs, VOLUME="TOCS11", NUMBER=1, PAGES="1-32", YEAR=1993, REFERENCES=30, KEYWORDS="Design; measurement; performance improvement", ABSTRACT="In this paper we propose a new paradigm, chores, for the run-time support of parallel computing on shared-memory multiprocessors. We consider specifically uniform memory access shared-memory environments, although the chore paradigm should also be appropriate for use within the cluster of a large-scale nonuniform memory access machine.", } @INPROCEEDINGS{Eng93:Memory, AUTHOR="K. Y. Eng and Mark Karol and R. D. Gitlin", TITLE="Memory and Channel-Sharing Techniques for Congestion Control in {ATM} Networks", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=1, PAGES="266-273", MONTH="March/April", YEAR=1993, REFERENCES=7, KEYWORDS="ATM; BISDN; Switching; Cross Connect; Central Buffer; Resource Sharing; Channel Assignment", ABSTRACT="We study the use of two resource sharing techniques for congestion control in a backbone ATM network consisting of cross-connects interconnected by high-speed optical links. Memory sharing allows an optimal shared use of buffer space in an ATM switch fabric. Channel sharing, on the other hand, takes advantage of spare line capacity in the system to permit temporary increases in the output rates of the congested output", } @INPROCEEDINGS{Evan93:Priority, AUTHOR="Joseph B. Evans and E. Duron and Yixin Wang", TITLE="Analysis and Implementation of a Priority Knockout Switch", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="1099-1106", MONTH="March/April", YEAR=1993, REFERENCES=14, KEYWORDS="ATM; Switching; Knockout Switch; Priority; Packet Discarding; Loss Probability; Performance Evaluation; nalysis; Simulation", ABSTRACT="A Priority Knockout Switch (PKS) architecture employing prioritybased congestion control mechanisms is presented in this paper. This architecture allows the capability to assign qualities of service to connections through per cell or virtual circuit priority mechanisms. Performance studies which use an analyticalmodel for uniform traffic and simulation models for uniform and bursty traffic are presented.", } @INPROCEEDINGS{Fabi93:Polling, AUTHOR="O. Fabian and H. M. Levy", TITLE="Polling System Optimization Through Dynamic Routing Policies", BOOKTITLE=infocom, ADDRESS="San Francisco", PAGES="194-200", MONTH="March/April", YEAR=1993, REFERENCES=12, KEYWORDS="Polling system; optimization; dynamic routing", ABSTRACT="Polling systems have been used as a central model for modeling various communications networks, like the token ring network, somd high speed local-area networks and others. In this paper we focus on pseudo cyclic algorithms which can prioritize the various stations by dynamically changing their service order but maintain fairness by visiting each station exactly once in a cycle.", } @INPROCEEDINGS{Flik93:Traffic, AUTHOR="Z. Flikop", TITLE="Traffic Management in Packet Switching Networks", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=1, PAGES="25-29", YEAR=1993, REFERENCES=4, KEYWORDS="Routing; decision theory; expert system; optimization; packet switching", ABSTRACT="This article discusses some problems in the development of routing management systems for packet switching networks. To decreasereaction time in routing management, we suggest the development of expert systems with control rules that do not require feedback. Considering the complexity of the development of the control rules, the use of self-learing together with a two-step optimization is suggested.", } @INPROCEEDINGS{Fons93:Estimating, AUTHOR="NELSON {L. S. DA FONSECA} and J. A. Silvester", TITLE="Estimating the loss probability in a multiplexer loaded with multi-priority {MMPP} streams", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=2, PAGES="1037-1041", YEAR=1993, REFERENCES=15, KEYWORDS="ATM; multiplexer; MMPP; loss probability; multiple priority", ABSTRACT="Future BISDN will carry a wide spectrum of applications with different loss requirements. By adopting a multi-priority schemeat the cell level, we gain flexibility in coping with these diverse requirements in an efficient way. In this paper, we address the issue of furnishing a multiclass selective discard mechanism. We compute the loss probability per class where each class is modelled as a two-state MMPP.", } @INPROCEEDINGS{Frie93:Effect, AUTHOR="V. J. Friesen and J. W. Wong", TITLE="The Effect of Multiplexing, Switching and Other Factors on the performance of Broadband Networks", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="1194-1203", MONTH="March/April", YEAR=1993, REFERENCES=28, KEYWORDS="ATM; statistical multiplexing; source policing; leaky bucket; bursty traffic; IPP; interrupted poisson process; MMPP; simulation; delay", ABSTRACT="A network model is developed to study the effect of multiplexingand switching on broadband network performance. Emphasis is placed on the queueing behaviour of the network, and the impact of multiplexing and switching on the burstiness of the various traffic streams. The three parameters commonly used to characterize traffic (namely peak bit rate, mean bit rate, and mean burst length) are also considered, and the dependence of", } @ARTICLE{Ghei93:CHAOS, AUTHOR="Ahmed Gheith and Karsten Schwan", TITLE="{CHAOS:} Kernel Support for Multiweight Objects, Invocations, and Atomicity in Real-Time Multiprocessor Applications", JOURNAL=tocs, VOLUME="TOCS11", NUMBER=1, PAGES="33-72", YEAR=1993, REFERENCES=62, KEYWORDS="Operating system; parallel software; measurement; design", ABSTRACT="CHAOS is an object-based multiprocessor operating system kernel that provides primitives with which programmers may easily construct objects of differing types and object invocations of differing semantics, targeting multiprocessor systems, and real-time applications. This paper reviews the primitives offered by CHAOS and demonstrates how the required elements of the CHAOS real-time kernel are constructed with those primitives.", } @INPROCEEDINGS{Gian93:Shared, AUTHOR="S. Gianatti and Achille Pattavina", TITLE="Performance Analysis of Shared-Buffered Banyan Networks under arbitrary Traffic Patterns", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="943-952", MONTH="March/April", YEAR=1993, REFERENCES=15, KEYWORDS="ATM; Switching; Banyan Network; Central Buffer; Bursty Traffic; Unbalanced Traffic; Performance Evaluation; Analysis", ABSTRACT="This paper addresses the performance study of multistage Banyan networks with a shared buffer in each switching element. Two different internal protocols are considered for the transfer of packets from stage to stage based on the presence or absence of interstage backpressure signals to signal the occurrence of buffer saturation conditions. Two kinds of traffic patterns havebeen studied, a bursty and an unbalanced traffic pattern.", } @INPROCEEDINGS{Hart93:User, AUTHOR="V. F. Hartanto and Harsha Sirisena", TITLE="User-Network Policer: A New Approach for {ATM} Congestion Control", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=1, PAGES="376-383", MONTH="March/April", YEAR=1993, REFERENCES=13, KEYWORDS="ATM; Policing Function", ABSTRACT="In this paper, we propose a new approach to the policing function called User-Network Policer. The scheme consists of a service-dependent User Policer and a service-independent NetworkPolicer. Users are responsible to mark and to police their traffic appropriately before sending them into the network, On the other hand, the network is only responsible to verify the correctness of user policing and to transport cells", } @INPROCEEDINGS{Heid93:Parallel, AUTHOR="P. Heidelberger and D. M. Nicol", TITLE="Parallel Algorithms for Simulating Continuous Time Markov Chains", BOOKTITLE="7th Workshop on PADS ACM - SIGSIM", ADDRESS="San Diego", VOLUME=23, PAGES="11-18", YEAR=1993, REFERENCES=15, KEYWORDS="Parallel algorithm; simulation; continuous time process; Markov process", ABSTRACT="We have previously shown that the mathematical technique of uniformization can serve as a basis of synchronization for the parallel simulation of continuous time Markov chains. This paperreviews the basic method and compares four different methods based on uniformaization, evaluating their strengths and weaknesses as a function of problem characteristics.", } @INPROCEEDINGS{Herr93:Correlation, AUTHOR="C. Herrmann", TITLE="Correlation effect on per-stream {QOS} parameters of {ATM} traffic superpositions relevant to connection admission control", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=2, PAGES="1027-1031", YEAR=1993, REFERENCES=14, KEYWORDS="ATM; connection admission control; QOS; DMAP; finite buffer", ABSTRACT="Connection Admission Control (CAC) is usually modelled by superimposing the existing traffic and the new connection on each ATM link. Correlation in the single streams can affect the loss probabilities and waiting times of both streams. In order to investigate this influence, this paper presents an exact solution of the DMAP1+DMAP2/D/1/s finite buffer queue. The analysis is done by means of the SMP[2]/D/1/s finite buffer queue.", } @INPROCEEDINGS{Herr93:Discrete, AUTHOR="C. Herrmann", TITLE="Analysis of the Discrete-Time {SMP/D/1/s} Finite Buffer Queue with Applications in {ATM}", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=1, PAGES="160-167", MONTH="March/April", YEAR=1993, REFERENCES=16, KEYWORDS="Semi Markov Process; Queueing System; Finite Capacity; Analysis; Loss Probability", ABSTRACT="This paper presents an exact analysis of the discrete-time single server SMP/D/1/s queue with an arbitrary Semi-Markov input process and a buffer size s. The SSMP/D/1/s queue as well as the DMAP/D/1/s queue are covered by this model. The analysis yields easy expressions for the queue length density at arrivals(loss probability) and the waiting time density. This queueing system is suitable for modelling correlated input streams of ATM", } @INPROCEEDINGS{Hluc93:Queueing, AUTHOR="M. G. Hluchyi and Nanying Yin", TITLE="On the Queueing Behaviour of Multiplexed Leaky Bucket Regulated Sources", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=2, PAGES="672-679", MONTH="March/April", YEAR=1993, REFERENCES=13, KEYWORDS="ATM; Source Policing; Policing Function; Leaky Bucket; Performance Evaluation", ABSTRACT="In this paper, we investigate the effectiveness of the unbuffered leaky bucket algorithm in protecting the quality of service experienced by connections passing through a common internodal link queue. We show that excess traffic generated by asource can pass through the Leaky Bucket without being marked ordiscarded, and cause unacceptable packet delay/loss to other connections. We suggest allocating network bandwidth above the", } @INPROCEEDINGS{Hsin93:Study, AUTHOR="D. K. Hsing", TITLE="Performance study on the ``Leaky Bucket'' Usage Parameter Control mechanism with {CLP} tagging", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=1, PAGES="359-364", YEAR=1993, REFERENCES=12, KEYWORDS="ATM; performance evaluation; leaky bucket", ABSTRACT="In this paper, it is attempted to quantify the value of using the option of setting an ATM cell header at the Usage Parameter Control (UPC) mechanism. We assume that the UPC mechanism uses a ``leaky bucket'' algorithm. The statistical gain realized in the ATM network with a UPC mechanism using CLP tagging is compared to that with a UPC mechanism not using CLP tagging.", } @INPROCEEDINGS{Jeon93:Achievability, AUTHOR="Y. H. Jeon and I. Viniotis", TITLE="Achievability of combined {GOS} Requirements in Broadband Networks", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=1, PAGES="192-196", YEAR=1993, REFERENCES=6, KEYWORDS="Queueing discipline; broadband; communication network; delay; loss probability", ABSTRACT="In the queueing literature, achievable performances for average delay and probability of loss have been traditionally referred to as conservation laws. The conservation laws presented so far, deal with single performance measures such as delay only orloss only, under a single arrival or service control. Our emphasis in this paper is on laws for combined arrival and service controls. We also present a dynamic queueing discipline.", } @INPROCEEDINGS{Jeon93:Achievable, AUTHOR="Y. H. Jeon and I. Viniotis", TITLE="Achievable Loss Probabilities and Buffer Allocation Policies in {ATM} Nodes with correlated arrivals", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=1, PAGES="365-369", YEAR=1993, REFERENCES=14, KEYWORDS="ATM; loss probability; buffer management; traffic control; MMPP", ABSTRACT="We consider the synthesis problem of a traffic control scheme based on conservation laws. Conservation laws, that characterizeachievable performance regions, have been very useful for analysis of delay and loss measures in systems, with iid arrivals. In this study, we derive a conservation law for cell loss probabilities, for an ATM node with a finite buffer size and correlated arrival processes.", } @INPROCEEDINGS{Jone93:Generic, AUTHOR="S. K. Jones and C. Smythe", TITLE="A Generic Framework for the Simulation Analysis of Protocol Layered Communication Systems", BOOKTITLE=mascots, ISBN="1-56555-018-8", ADDRESS="San Diego", PAGES="183-186", YEAR=1993, REFERENCES=11, KEYWORDS="Simulation; object-oriented programming; ISO reference model", ABSTRACT="A communications network performance modelling methodology based on the specification of a generic network architecture is proposed. This methodology provides a framework upon which the analytic, simulation and formal analysis of protocol layered communications systems can be performed. This paper concentrates on the application of ths framework to simulation modelling.", } @INPROCEEDINGS{Jou93:Tractable, AUTHOR="Y. F. Jou and F.-Y. Lai and A. Nilsson", TITLE="Tractable Analysis of a finite capacity polling system under bursty and correlated {ATM} arrivals", BOOKTITLE=icc, ADDRESS="Geneva, Switzerland", VOLUME=1, PAGES="340-344", YEAR=1993, REFERENCES=6, KEYWORDS="ATM; MMBP; mean delay; loss probability; finite capacity; polling system; bursty traffic", ABSTRACT="This paper is concerned with the mean delay and the cell loss probability that bursty and correlated arrivals incur in an ATM switching module which can be modelled as a finite capacity polling system with nonexhaustive cyclic service. The arrival process to each input port of the system is modelled by a Markov Modulated Bernoulli Process (MMBP) which is a able to describe the bursty and correlated nature of the ATM traffic.", } @INPROCEEDINGS{Jung93:Translation, AUTHOR="J. Jung and D. Seret", TITLE="Translation of {QoS} Parameters into {ATM} Performance Parameters in {BISDN}", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=2, PAGES="748-755", MONTH="March/April", YEAR=1993, REFERENCES=25, KEYWORDS="ATM; QOS; Layered Protocol", ABSTRACT="The main goal of this paper is to present the translation of QoSparameters between layers. We concentrate on the translation from the AAL to the ATM layer because the AAL must be designed to be service dependent and specific. The QoS parameters at the ATM layer resulting from the translation will be the performancerequirements on a connection basis for the ATM network. Finally, we discuss the impacts of QoS parameter translation on", } @INPROCEEDINGS{Kala93:Fast, AUTHOR="N. Kalantery and S. C. Winter and D. R. Wilson and A. P. Redfern", TITLE="Fast Parallel Simulation of {SS7} Telecommunication Networks", BOOKTITLE=mascots, ISBN="1-56555-018-8", ADDRESS="San Diego, California", PAGES="171-175", NOTE="SCS Simulation Series", YEAR=1993, REFERENCES=9, KEYWORDS="Distributed simulation; parallel simulation; SS7; signaling", ABSTRACT="This paper describes a basic parallel simulation model for SS7 telecommunication networks. Experimental results on the performance of the parallel simulator are discussed. Based on the current results and by pointing out the special lookahead characteristics of the SS7 networks, a technique for achieving improved speedup of execution is proposed.", } @INPROCEEDINGS{Kang93:Queueing, AUTHOR="Cheoul-Shin Kang and H. H. Tan", TITLE="Queueing Analysis of Explicit Priority Assignment Partial Buffer Sharing Schemes for {ATM} Networks", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=2, PAGES="810-819", MONTH="March/April", YEAR=1993, REFERENCES=11, KEYWORDS="ATM; Multiple Priority; Buffer Management; Performance Evaluation; Analysis; Bursty Traffic; Loss Probability", ABSTRACT="This paper considers the explicit assignment space priority control policy employing the partial buffer sharing scheme. Cellloss probabilities are derived for a traffic model consisting oftwo service classes. Each service class consists of bursty traffic generated by a multiple number of three-state discrete-time Markov sources. The effect of cell loss probabilities on traffic load and burstiness characteristics, as", } @INPROCEEDINGS{Kani93:Copy, AUTHOR="Jaidev Kaniyil and Y. Onozato and W. D. Zhong", TITLE="A Copy Network with Shared Buffers for Large Multicast {ATM} switches", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=2, PAGES="722-726", YEAR=1993, REFERENCES=16, KEYWORDS="ATM; buffer management; multicast; Banyan network; nonblocking; performance analysis; architecture", ABSTRACT="This paper proposes a new architecture for the copy network which is an integral part of multicast ATM switches. The new architecture makes use of the property that the Broadcast BanyanNetwork (BBN) is non-blocking if the active inputs are cyclically concentrated and the outputs are monotone. In the new architecture, by employing a token ring reservation scheme, the outputs of the copy net. are reserved before multic. cell is replicated.", } @INPROCEEDINGS{Kim93:Multinet, AUTHOR="H. Kim", TITLE="Multinet Switch: Multistage {ATM} Switch Architecture with PartiallyShared Buffers", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=2, PAGES="473-480", MONTH="March/April", YEAR=1993, REFERENCES=16, KEYWORDS="ATM; Switching Network; Architecture; Self Routing; Multistage Interconnection Network; Multicast", ABSTRACT="The proposed switch, Multinet switch, is a self-routing multistage switch with internal buffers capable of achieving 100\% throughput. Although it provides incoming ATM cells with multiple paths, the cell sequence is maintained throughout the switch fabric thus eliminating the out-of-order cell sequencing problem. Cells contending for the same output addresses are buffered internally according to a partial buffer sharing", } @INPROCEEDINGS{Kim93:Policing, AUTHOR="Jaehyung Kim and D. Y. Kwak and H. S. Park and W. S. Rhee", TITLE="The policing mechanism with state-dependent control", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=2, PAGES="1042-1046", YEAR=1993, REFERENCES=11, KEYWORDS="ATM; policing function; traffic shaping", ABSTRACT="In this paper, we propose the policing mechanism which can control the buffer read rate properly according to the buffer state. As this mechanism has the buffer with threshold, it has intrinsically traffic shaping function nd can control the bursty trafficflow to a some extent. Simulation results show that our model can easily manage the cell loss probability for a given buffer size by controlling the read rate according to the buffer state.", } @ARTICLE{Ko93:Two, AUTHOR="C. C. Ko and Wai-Choong Wong and Jianxuan Du and R. H. Deng and K. M. Lye", TITLE="Performance analysis of two bridged {CSMA/CD} networks", JOURNAL=ccr, VOLUME=16, NUMBER=8, PAGES="501-510", YEAR=1993, REFERENCES=22, KEYWORDS="Csma/cd; bridge; performance analysis", ABSTRACT="This paper analyses the performance of two non-persistent CSMA/CD LANs linked by a bridge. The main function of the bridge is to buffer all internetwork packets and forward them across networks to their destinations. From establishing and solving the main equations governing the behaviour of the bridge networks, the effects of design parameters such as the bridge buffer size and re-transmission backoff delay on the performance are studied.", } @INPROCEEDINGS{Koba93:Diffusion, AUTHOR="H. Kobayashi and Q. Ren", TITLE="A diffusion approximation analysis of an {ATM} statistical multiplexer with multiple types of traffic part {I:} equilibrium state solutions", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=2, PAGES="1047-1053", YEAR=1993, REFERENCES=23, KEYWORDS="ATM; multiplexer; diffusion approximation; statistical multiplexer", ABSTRACT="We introduce a multi-dimensional diffusion model to characterizethe ``on-off'' source behaviour in an ATM statistical multiplexer, where multiple types of traffic are concentrated. Under a reasonable set of assumptions, this diffusion process can then be approximated by a multi-dimensional Ornstein-Uhlenbeck process, which is a Gaussian Markov process. The packet arrival process is shown to be a Gaussian (but not Markov) process.", } @INPROCEEDINGS{Kokk93:Optimal, AUTHOR="M. Logothetis and Shigeo Shioda and G. Kokkinakis", TITLE="Optimal Virtual Path Bandwidth Management assuring Network reliability", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=1, PAGES="30-36", YEAR=1993, REFERENCES=17, KEYWORDS="ATM; bandwidth allocation; virtual path; BISDN; reliability", ABSTRACT="The impact of virtual path bandwidth (VPB) management on the network performance in the expected environment of BISDN has been recognized and several VPB control schemes have been proposed. We present the optimal VPB management which minimizes the worst call blocking probability of all VPs in the network. We solve a large network optimization problem by a rigorous analytical procedure; bandwidth distribution; assuring reliability", } @INPROCEEDINGS{Kris93:There, AUTHOR="Ram Krishnan and N. F. Maxemchuk", TITLE="Is There Life Beyond Linear Topologies? A Comparison of {DQDB} and Manhattan Street Network", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=1, PAGES="690-698", MONTH="March/April", YEAR=1993, REFERENCES=12, KEYWORDS="Man; DQDB; comparison", ABSTRACT="In this paper, we make a performance and reliability comparison of two networks that have been proposed for use as high speed Metropolitan Area Networks (MAN's) - the Manhattan Street network and the DQDB network. We show that the MSN provides a much higher network throughput for a variety of traffic patterns - uniform and non-uniform. We also look at the reliabilitz of both networks and show that MSN can survive more failures than DQDB.", } @INPROCEEDINGS{Kwon93:Simple, AUTHOR="Hyuck Kwon and A. Tubtiang and", TITLE="A Simple Flow Control Mechanism in {ATM} Network with End to End transport", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=2, PAGES="654-661", MONTH="March/April", YEAR=1993, REFERENCES=21, KEYWORDS="ATM; congestion Control; end-to-end flow control; window mechanism", ABSTRACT="In traditional data networks, the window flow control combined with error control was used prevalently. But, the window flow control mechanism might be useless in ATM networks because the propagation delay is too large compared with the transmission rate. In this paper, we propose a simple flow control mechanism, called RCT (Rate Control for end-to-end Transport), for end-to-end data transport. The RCT shows acceptable performance", } @INPROCEEDINGS{LaPo93:MSP, AUTHOR="Thomas La Porta and M. Schwartz", TITLE="Performance Analysis of {MSP:} A Feature-Rich High-Speed Transport Protocol", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=1, PAGES="513-520", MONTH="March/April", YEAR=1993, REFERENCES=9, KEYWORDS="Transport protocol; performance analysis; multiprocessor", ABSTRACT="This paper presents a performance analysis of the MultiStream Protocol (MSP), a feature-rich flexible transport protocol designed to meet the needs of high-performance applications. The analysis is based on the processing complexity associated with performing certain protocol functions, and highlights the advantages and disadvantages of the implementation of several different protocol mechanisms, such as packet or block mode data transfer.", } @INPROCEEDINGS{Lau93:Traffic, AUTHOR="Wing Cheong Lau and Sanqi Li", TITLE="Traffic Analysis in Large-Scale High-Speed Integrated Networks: Validation of Nodal Decomposition Approach", BOOKTITLE=infocom, ADDRESS="San Francisco", PAGES="1320-1329", YEAR=1993, LANGUAGE="English", REFERENCES=9, KEYWORDS="ATM; statistical multiplexer; output process; simulation; bursty traffic; MMBP", ABSTRACT="In this paper we study the changes in individual source characteristics as traffic flows through a network. The effects of inter-source dependence caused by multiplexing at various network nodes are also investigated. Each traffic source is modeled as a 2-state or multiple-state Markov modulated process. Extensive simulations have been performed for homogeneous and heterogeneous traffic environments. Various network configurations have also been considered. We have identified the factors and parameters which have dominant effects on source distortions and inter-source cross-correlations. We also discuss the implications of our findings on the feasibility of decomposing the analysis of an ATM network into a set of separate nodal analyses.", ANNOTE="In this paper we determine the conditions under which nodal decompositions can be applied for network-wide, multimedia traffic analysis. Through extensive simulation studies on individual departure source characteristics and inter-source cross-correlation at the output side of a network node, the nodal decomposition approach has been validated for large-scale, high-speed integrated networks.", URL="http://www.ece.utexas.edu/~sanqi/papers.html", } @INPROCEEDINGS{Lee93:ATM, AUTHOR="Hwang-Soo Lee and J. W. Mark", TITLE="{ATM} Network Traffic Characterization Using Two Types of On-Off Sources", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=1, PAGES="152-159", MONTH="March/April", YEAR=1993, REFERENCES=12, KEYWORDS="Statistical Multiplexer; Bursty Traffic; MMPP; Correlated Variate; Approximation; Performance Evaluation; Finite Queue; Loss Probability", ABSTRACT="An approximate model is considered for characterizing aggregatedcell arrivals from heterogeneous sources which may generate cells in a highly time correlated manner. The model consists of two types of on-off sources and provides enough degrees of freedom to match a number of moments of cell arrival rates as well as the time correlation of the aggregated cell arrivals. Numerical examples are presented to illustrate the accuracy", } @INPROCEEDINGS{Leon93:StarBurst, AUTHOR="A. Leon-Garcia and I. Widjaja", TITLE="The StarBurst Switch with dedicated Output Buffering and Shared input Buffering", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=2, PAGES="691-695", YEAR=1993, REFERENCES=7, KEYWORDS="ATM; performance evaluation; input buffer; output buffer; bursty traffic", ABSTRACT="The performance of an ATM switch can be seriously degraded undersome traffic types. The StarBurst architecture is a class of ATM switches which can employ both dedicated buffering and shared buffering. We show that the switch capacity can achieve that of output queueing switches.", } @INPROCEEDINGS{Li93:Fundamental, AUTHOR="Song Chong and Song Li", TITLE="Fundamental Limits of Input Rate Control in High Speed Network", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=2, PAGES="662-671", MONTH="March/April", YEAR=1993, REFERENCES=15, KEYWORDS="ATM; source policing; leaky bucket; analysis; fluid flow model; spectral analysis", ABSTRACT="In this paper we study the performance trade-off of access rate control in high speed networks by spectral analysis. The emphasis is placed on the interrelationship between access control queueing and network queueing, as affected by degrees of control in response to differing input power spectrum. Both deterministic and stochastic analyses are developed. While the simple deterministic analysis helps us to gain a good engineering insight into access rate control in high speed networks, the stochastic analysis provides us quantitative measures of the access rate control in random traffic environment. Interesting observations are made between control performance and low-frequency input statistics. In the stochastic analysis we are able to derive the exact solutions of a network queue fed by a correlated source under the leaky-bucket access control.", ANNOTE="In this paper we explore the fundamental limits of input rate control by spectral analysis in frequency domain. Both deterministic and stochastic analyses are developed. Especially, the simple deterministic analysis helps us to gain a great knowledge of performance trade-off for input rate control in high speed network.", URL="http://www.ece.utexas.edu/~sanqi/papers.html", } @INPROCEEDINGS{Liew93:Fundamental, AUTHOR="S. C. Liew and T. T. Lee", TITLE="A Fundamental Property for Traffic Management in {ATM} Networks", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="1240-1249", MONTH="March/April", YEAR=1993, REFERENCES=14, KEYWORDS="ATM; Switching; Traffic Control", ABSTRACT="It is desirable that the traffic-control policy at a network node depends only on the external traffic loads on the input and output links, but not on the detailed addressing or distributionof packets from inputs to outputs. It should be possible to guarantee the grade-of-service of an input-output connection by controlling the aggregate loads on the input and output. Switch nodes in which such a traffic-control policy is possible are", } @INPROCEEDINGS{Lin93:Minimax, AUTHOR="F. Lin and Jin-Tuu Wang", TITLE="Minimax Open Shortest Path First Routing Algorithms in Networks supporting the {SMDS} Service", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=2, PAGES="666-670", YEAR=1993, REFERENCES=15, KEYWORDS="Shortest path; routing algorithm; SMDS", ABSTRACT="In this paper, we present and compare two quasi-static minimax Open Shortest Path First (OSPF) routing algorithms in networks supporting the SMDS service. The emphasis of this paper is (i)to consider a quasi-static ISSI routing scheme and present a formal problem formulation and (ii) to develop an efficient and effective algorithm to solve the problem.", } @INPROCEEDINGS{Logo93:Reliability, AUTHOR="D. Logothetis and K. S. Trivedi", TITLE="Reliability Analysis of Various Station Attachment Schemes in a {FDDI} Token Ring", BOOKTITLE=infocom, ADDRESS="San Francisco", PAGES="201-208", MONTH="March/April", YEAR=1993, REFERENCES=7, KEYWORDS="FDDI; reliability; MAN; hslan", ABSTRACT="Five different attachment schemes proposed for the FDDI token ring are compared in terms of reliability. For this purpose, the topologies are first studied in isolation (reliability of the path to the backbone) and subsequently end-to-end user relibailities are compared by combining backbone relibaility wiht the relibaility of the path to the backbone.", } @INPROCEEDINGS{Mark93:Computational, AUTHOR="J. W. Mark and G. Wu", TITLE="Computational methods for performance evaluation of an {ATM} multiplexer supporting bursty traffic", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=2, PAGES="1022-1026", YEAR=1993, REFERENCES=12, KEYWORDS="ATM; performance evaluation; bursty traffic; multiplexer", ABSTRACT="An ATM multiplexer supporting a number of bursty sources is modeled as a discrete time, single server queueing system with an infinite buffer. The probability generating function (PGF) method is used to analyze the queueing behavior. The PGF method requires a large number of boundary values, and hence the roots of the characteristic equation. An iterative algorithm for evaluating the roots is proposed.", } @INPROCEEDINGS{Mase93:Fiber, AUTHOR="F. Masetti and P. Gavignet-Morin and D. Chiaroni and G. Da Loura", TITLE="Fiber Delay Lines Optical Buffer for {ATM} Photonic Switching Applications", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="935-942", MONTH="March/April", YEAR=1993, REFERENCES=46, KEYWORDS="ATM; optical switching; buffer; implementation; WDM; wavelength division multiplex", ABSTRACT="This paper presents an optical buffer shift register based on a set of fiber delay lines, with the goal of providing cell storage and queueing in ATM photonic switches. The concept of multi-wavelength cell buffering is adopted, to realise a shared buffer with a storage capacity equivalent to a set of shift registers. Since one wavelength-encoded cell only must enter onedelay line at each time slot, contention resolution is necessary", } @ARTICLE{Mate93:Proposal, AUTHOR="F. Matera and E. Ripani and M. Romagnoli and M. Settembre", TITLE="Proposal of an all Optical Shuffle Multihop Network", JOURNAL=ett, ADDRESS="Milano", VOLUME=4, NUMBER=2, PAGES="213-219", YEAR=1993, REFERENCES=13, KEYWORDS="Optical LAN; optical switching; hsLAN; MAN; shufflenet", ABSTRACT="In this paper a shuffle multihop network, with a routing and a processing completely obtained at optical level, is proposed for the first time. Such a network can be performed by means of optical AND gates based on second harmonic generation and optical XOR gates based on soliton dragging process. Soliton pulses are used to avoid the distortion caused by chromatic despersion. The main characteristics of the network: 64 nodes, 50 Gbit/s", } @INPROCEEDINGS{McAu93:Fast, AUTHOR="Anthony McAuley and P. Francis", TITLE="Fast Routing Table Lookup Using {CAMs}", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="1382-1391", MONTH="March/April", YEAR=1993, REFERENCES=9, KEYWORDS="Hierarchical Network; Routing Table; Hierarchical Routing; Implementation", ABSTRACT="This paper investigates fast routing table lookup techniques, where the table is composed of hierarchical addresses such as those found in a national telephone network. The hierarchical addresses provide important benefits in large networks; but existing fast routing table lookup techniques, based on hardwaresuch as Content Addressable Memory (CAM), work only with flat addresses. We present several fast routing table lookup", } @INPROCEEDINGS{Meem93:New, AUTHOR="G. Meempat and G. Ramamurthy and B. Sengupta", TITLE="A New Performance Measure for Statistical Multiplexing: Perspective of the Individual Source", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=2, PAGES="531-538", MONTH="March/April", YEAR=1993, REFERENCES=12, KEYWORDS="ATM; Fast Packet Switching; Service Degradation; Performance Measure; Loss Calculation; Analysis; Fluid Flow Model", ABSTRACT="In this paper, we introduce the notion of perceptible degradation of quality of the received signal seen by an individual user. We propose that the design of high-speed networks that support real-time traffic be carried out on the basis of a new performance measure, which is defined as the expected number of perceptible degradations that a customer is likely to encounter per unit of time. By using fluid-flow", } @INPROCEEDINGS{Mera93:Priority, AUTHOR="L. Merakos and L. K. Reiss", TITLE="Priority Shaping of Source Traffic for {ATM} networks", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=2, PAGES="671-675", YEAR=1993, REFERENCES=4, KEYWORDS="ATM; traffic shaping; priority", ABSTRACT="Priority shaping of aggregate source traffic is proposed for usewith policed virtual paths in ATM networks. The shaping functionlocated between the traffic sources sharing the path and the network edge, acts as a self-policing mechanism augmented by a form of priority service. We propose two versions of the shaperand discuss their interaction with a virtual path policed by a leaky bucket mechanism optimization is considered.", } @INPROCEEDINGS{Merc93:Characterization, AUTHOR="G. Mercankosk and Antonio Cantoni", TITLE="Characterization of a {CBR} Connection over a Channel with Known Bounded Delay Variation", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="1170-1177", MONTH="March/April", YEAR=1993, REFERENCES=7, KEYWORDS="Buffer Dimensioning; Delay Constraint; Reassembly Delay; Periodic Input Traffic", ABSTRACT="In this paper, we examine how to provide constant bit rate service over a channel that can be characterised by a constant transmission delay plus a variable cell access delay. To compensate for the variability in the access delay, buffers are used at the source and at the destination ends of the connection. In the paper, we determine the minimum size of buffers required at the transmitter and the receiver to provide", } @INPROCEEDINGS{Mill9303:Dynamic, AUTHOR="G. J. Miller and M. Paterakis", TITLE="Performance Evaluation of a Dynamic Priority mechanism for the pi-Persistent Protocol for Metropolitan Area Networks", BOOKTITLE=infocom, ADDRESS="San Francisco", PAGES="707-714", MONTH="March/April", YEAR=1993, REFERENCES=15, KEYWORDS="Persistent; MAN; performance evaluation; dynamic priority", ABSTRACT="In this paper, we present a dynamic control algorithm for the multi-priority pi-persistent protocol that allows the protocol to adapt to changing network conditions in a fully distributed manner. Through a simulation model, we study the performance of the dynamic algorithm under various changing network scenarios, and we demonstrate the ability of the protocol to provide the desired quality of service to each priority class of traffic.", } @INPROCEEDINGS{Mouf93:Bridged, AUTHOR="Hussein Mouftah and R. Zarour", TITLE="Bridged Shuffle-Exchange Network: a high-perormance self-routing {ATM} switch", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=2, PAGES="696-700", YEAR=1993, REFERENCES=7, KEYWORDS="ATM; self routing; multistage interconnection network", ABSTRACT="A high-performance self-routing switch architecture, based on multistage interconnection network is proposed for ATM switchingsystems. It makes use of the principle of error-prevention routing. The switch performance is studied under uniform trafficand full load conditions.", } @INPROCEEDINGS{Ayan93:Fast, AUTHOR="Ender Ayanoglu", TITLE="A fast topology update algorithm for restoration under multiple failures in broadband networks", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=3, PAGES="1295-1299", YEAR=1993, REFERENCES=2, KEYWORDS="ATM; broadband; failure; topology", ABSTRACT="In this paper, we propose a topology update algorithm, a particular flooding algorithm, against multiple link, path, or nodefailures. The algorithm accomplishes restoration faster than a similar algorithm proposed by up to three times for the first failure, up to five times for the secobd failure, up to 2n+1 timesfor the nth failure. The algorithm simplifies the protocol used in restoration, and reduces the number of messages passed in net", } @ARTICLE{Nico93:Cost, AUTHOR="D. M. Nicol", TITLE="The Cost of Conservative Synchronization in Parallel Discrete Event Simulations", JOURNAL=jacm, VOLUME=40, NUMBER=2, PAGES="304-333", YEAR=1993, REFERENCES=32, KEYWORDS="Distributed simulation; performance analysis; algorithm", ABSTRACT="This paper analytically studies the performance of a synchronous conservative parallel discrete-event simulation protocol. The class of models considered simulates activity in a physical domain, and possesses a limited ability to predict future behaviour.", } @INPROCEEDINGS{Nico93:Global, AUTHOR="D. M. Nicol", TITLE="Global synchronization for optimistic parallel discrete event simulation", BOOKTITLE="7th Workshop on PADS ACM - SIGSIM", ADDRESS="San Diego", VOLUME=23, PAGES="27-34", YEAR=1993, REFERENCES=10, KEYWORDS="Discrete event simulation; parallel algorithm; simulation; synchronization", ABSTRACT="In this paper we describe an efficient parallel algorithm derived from a common barrier synchronization algorithm used in parallel processing. The algorithm's principle attraction is speed, and generality - it is designed to be used in contexts more general than parallel discrete-event simulation. To establish our claim to speed, we compare our algorithm's performance with the standard barrier algorithm", } @INPROCEEDINGS{Ogie93:Efficient, AUTHOR="R. G. Ogier and Vlad Rutenburg", TITLE="Efficient Algorithms for optimal alternate Routing in communication networks", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=2, PAGES="676-680", YEAR=1993, REFERENCES=9, KEYWORDS="Alternate routing; communication network; shortest path", ABSTRACT="Three new efficient distributed and centralized algorithms are presented for precomputing shortest alternate paths in a communication network. Such alternate paths can be used in case one or more links or nodes fail on the primary path between any two nodes.", } @INPROCEEDINGS{Ohta9303:Peak, AUTHOR="C. Ohta and H. Tode and Makoto Yamamoto and H. Okada and Y. Tezuka", TITLE="Peak Rate Regulation Scheme for {ATM} Networks and its Performance", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=2, PAGES="680-689", MONTH="March/April", YEAR=1993, REFERENCES=12, KEYWORDS="ATM; Source Policing; Policing Function; Analysis; Matrix Analytic Method", ABSTRACT="In this paper, we address the rate control to restrict the peak cell rate of traffic. We study a rate control which can restrictthe number of cells during any fixed length time interval in a discrete-time environment. The scheme is realized by means of simple counter accumulation per slot, and it is easy to be synchronized with slots. We analyze this rate control scheme by using Matrix Analytic Techniques in order to investigate the", } @INPROCEEDINGS{Oie9303:Internally, AUTHOR="Yuji Oie and K. Kawahara and M. Murata and H. Miyahara", TITLE="Performance Analysis of Internally Unbuffered Large Scale {ATM} Switch with Bursty Traffic", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="1270-1279", MONTH="March/April", YEAR=1993, REFERENCES=9, KEYWORDS="ATM; switching; analysis; loss probability", ABSTRACT="In this paper, we treat a 3-stage switching configuration with no internal buffers; i.e., bufferless switches are employed at the first and second stages, and output-buffered switches at thethird stage. A short-term cell loss probability is analyzed in order to examine the influence of bursty traffic on performance of the bufferless switch used at the first two stages. Furthermore, we propose a 4-stage switching configuration with", } @INPROCEEDINGS{Panc9303:Bandwidth, AUTHOR="P. Pancha and Magda El Zarki", TITLE="Bandwidth Requirements of Variable Bit Rate {MPEG} Sources in {ATM} Networks", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="902-909", MONTH="March/April", YEAR=1993, REFERENCES=7, KEYWORDS="Packet Video; ATM; Bandwidth Allocation; Dynamic Resource Allocation; Traffic Prediction; Simulation; oss Probability", ABSTRACT="The complex traffic characteristics of VBR video sources makes them difficult to accommodate in ATM networks. To efficiently transport these services will require both an understanding of the source traffic and novel network control schemes. We study the performance of variable bandwidth allocation schemes which estimate the required bandwidth for a VBR MPEG source using a simple prediction algorithm.", } @INPROCEEDINGS{Petr93:End, AUTHOR="David Petr and J. S. Wineinger", TITLE="End-to-End Priority Cell Discarding Analysis for {ATM} Networks", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="1178-1185", MONTH="March/April", YEAR=1993, REFERENCES=10, KEYWORDS="ATM; Buffer Management; Packet Discarding; Priority; Performance Evaluation; Analysis; Loss Probability", ABSTRACT="We present extensions to a previous single-queue analysis that allow the calculation of end-to-end discarding probabilities foreach priority. We describe a computer tool based on this analysis that facilitates the calculation of end-to-end discarding probabilities for very general network topologies. Wecompare the computational efficiency of the computer-aided analytical tool with simulation techniques.", } @INPROCEEDINGS{Plot93:Parallel, AUTHOR="N. T. Plotkin and P. Varaiya", TITLE="Performance Analysis of Parallel {ATM} Connections for Gigabit SpeedApplications", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="1186-1193", MONTH="March/April", YEAR=1993, REFERENCES=8, KEYWORDS="Host; Interconnection; ATM; Virtual Channel; Traffic Distribution; Resequencing; Analysis; Bursty Traffic", ABSTRACT="We study a system which uses multiple ATM virtual circuits operating in parallel in order to connect two hosts at gigabit speeds. Packets in parallel channels can bypass each other, so reordering of packets before delivery to the host is required. This paper analyses performance parameters of this system including ATM channel delay, packet loss, and resequencing delay.", } @INPROCEEDINGS{Pope93:Modeling, AUTHOR="A. Popescu and Y. Ismailov and H. Rajaei and R. Ayani", TITLE="Modeling and Performance evaluation of Multiaccess Mechanisms at {SUPERLAN}", BOOKTITLE=mascots, ISBN="1-56555-018-8", ADDRESS="San Diego", PAGES="176-182", NOTE="SCS Simulation Series", YEAR=1993, REFERENCES=16, KEYWORDS="HSLAN; LAN; MAN; media access control", ABSTRACT="An issue of highest interest in the field of optical networks is to develop viable network architectures targeted to deal with the electrooptic bottleneck problem. A new multi-wavelength-dedicated-to-application concept, called SUPERLAN, has been proposed to solve this problem. This is an architecture where time-synchronous channels placed in different wavelengths are dedicated to different applications and control mechanisms.", } @INPROCEEDINGS{Pott93:Shared, AUTHOR="P. G. Potter and L. Wedding and Li-jun Yao and Moshe Zukerman", TITLE="A shared medium multi-service protocol", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=2, PAGES="1000-1004", YEAR=1993, REFERENCES=7, KEYWORDS="ATM; traffic shaping; protocol; priority", ABSTRACT="This paper describes a new ATM shared medium protocol which guarantees capacity for CBR and VBR services. The protocol includes a traffic shaping function, multiple prioritized access, a buffer insertion mechanism, destination release and distributed queueing for lower priority traffic. The protocol also allows for mixed priority mode for a connection to provide efficient and fair access for VBR connec. requiring a guaran. rate lower than peak.", } @INPROCEEDINGS{Rama93:Predictive, AUTHOR="G. Ramamurthy and B. Sengupta", TITLE="A Predictive Hop-by-Hop Congestion Control Policy for High-Speed Networks", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="1033-1041", MONTH="March/April", YEAR=1993, REFERENCES=15, KEYWORDS="ATM; Congestion Control; Traffic Prediction; Autoregressive Process; Performance Evaluation; Simulation", ABSTRACT="In this paper, we present a predictive congestion control policyfor carrying data traffic in the presence of real time traffic which have stringent delay requirements such as voice and video traffic. The policy is based on information exchange between adjacent nodes of the network. We characterise the cross trafficby its mean, its variance and the correlation, and we study the performance of our scheme over a wide range of cross traffic", } @ARTICLE{Rama93:Commercial, AUTHOR="G. Ramanathan and J. Oren", TITLE="Survey of Commercial Parallel Machines", JOURNAL=acmcan, VOLUME=21, NUMBER=3, PAGES="13-33", YEAR=1993, REFERENCES=16, KEYWORDS="Parallel computing; distributed computing system; survey", ABSTRACT="We have presented in this paper the survey of the parallel machines that are marketed today. The survey includes the latest machines available from Kendell Square Research, Thinking Machines Corporation, MasPar computer Corporation, NCUBE Corporation, Sequent Computer Systems and Parsytec. We have provided the topology, architecture, cache coherence, synchronization and performance in MFLOPs for each of the machines.", } @INPROCEEDINGS{Rein93:Statistical, AUTHOR="D. Reininger and others", TITLE="Statistical Multiplexing of {VBR} {MPEG} Compressed Video on {ATM} Networks", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="919-926", MONTH="March/April", YEAR=1993, REFERENCES=16, KEYWORDS="Packet Video; ATM; Traffic Source Model; Simulation; Loss Probability; Statistical Multiplexing", ANNOTE="This paper introduces a VBR MPEG video compression encoder , andevaluates the performance of a statistically multiplexed ATM network supporting a number of such VBR video sources. The results confirm that good ATM channel efficiencies (approximately 80-90\%) canbe obtained at reasonable cell loss rate and delay levels in thestatistical multiplexing scenario under consideration.", } @INPROCEEDINGS{Reis93:Shaping, AUTHOR="L. K. Reiss and L. Merakos", TITLE="Shaping of Virtual Path Traffic for {ATM} {BISDN}", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=1, PAGES="168-175", MONTH="March/April", YEAR=1993, REFERENCES=14, KEYWORDS="ATM; Virtual Path; Policing Function; Leaky Bucket; Analysis; Fluid Flow Model", ABSTRACT="Two mechanisms for shaping private network traffic in ATM BISDNare proposed and analyzed. The shapers utilize the information about the network policer and implement a form of priority service to assure satisfactory levels of cell loss and delay fortraffic streams with different service requirements multiplexed into a policed ATM virtual path. The performance of the shapers is analyzed using stochastic-fluid methods and Markov-modulated", } @INPROCEEDINGS{Rubi93:Burst, AUTHOR="I. Rubin and Kuang-Shin Lin", TITLE="A Burst-Level Adaptive Input-Rate Flow Control Scheme", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=2, PAGES="386-394", MONTH="March/April", YEAR=1993, REFERENCES=16, KEYWORDS="ATM; Congestion Control; Flow Control; Access Control; Performance Evaluation; Analysis", ABSTRACT="We propose a new input rate flow control scheme wherein the credit increment rate is updated periodically as the loading status varies. Based upon the observed status of each station's burst-level activity, the network access node distributes feedback control messages to the stations. These signalling messages allow the stations to adapt their credit increment rates in accordance with system burst-loading conditions. We", } @ARTICLE{Schm93:Transport, AUTHOR="Douglas Schmidt and T. Suda", TITLE="Transport System Architecture Services for High-Performance Communications Systems", JOURNAL=ieeejsac, VOLUME="SAC-11", NUMBER=4, PAGES="489-506", YEAR=1993, REFERENCES=46, KEYWORDS="Transport protocol; unix", ABSTRACT="This paper describes and classifies transport system services that integrate operating system resources such as CPU's, virtual memory, and I/O devices together with network protocols to support distributed multimedia applications running on local and wide area networks. A taxonomy is presented that compares and evaluates for commercial and experimental transport systems in terms of their protocol processing support.", } @INPROCEEDINGS{Shar93:Approximate, AUTHOR="Sanjay Kumar Sharma and D. Tipper", TITLE="Approximate Models for the Study of nonstationary Queues and their Applications to Communication Networks", BOOKTITLE=icc, ADDRESS="Geneva", VOLUME=1, PAGES="352-358", YEAR=1993, REFERENCES=13, KEYWORDS="ATM; Markovian queueing system; priority; M/D/1; fluid flow model; nonpreemptive priority", ABSTRACT="The modelling of ATM networks, and the effect of nonstationary traffic patterns, is of considerable interest to research community. In this paper, we present numerical methods based techniques to model multi-class Markovian traffic, with and without priority, in ATM networks. The use of these models is validated, under stationary and nonstationary traffic patterns, with simulation.", } @INPROCEEDINGS{Shen93:Second, AUTHOR="Hong-Dah Sheng and Song Li", TITLE="Second Order Effect of Binary Sources on Characteristics of Queue and Loss Rate", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=1, PAGES="18-27", MONTH="March/April", YEAR=1993, REFERENCES=20, KEYWORDS="Statistical Multiplexer; MMPP/G/1; Bursty Traffic; Correlated Variate; Spectral Analysis; Performance Evaluation; Finite Queue; Loss Probability", ABSTRACT="In this paper we characterize the second-order properties of a binary source by multi-state MMPP. A comprehensive numerical study is carried out to identify the individual effect of sourcesecond-order dynamics on queue length and loss rate. The conceptof input power spectrum is then developed as a unified source measurement for multimedia traffic analyses. This study well explores the inter-relationship between the source second-order", } @INPROCEEDINGS{Siba93:Class, AUTHOR="S. Sibal and Ji Zhang", TITLE="On a Class of Banyan Networks and Tandem Banyan Switching Fabrics", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=2, PAGES="481-488", MONTH="March/April", YEAR=1993, REFERENCES=9, KEYWORDS="ATM; Switching Network; Banyan Network; Multistage Interconnection Network; Output Buffer; Performance Evaluation; Simulation", ABSTRACT="This paper examines the operating characteristics of a class of Tandem Banyan Switching Fabrics (TBSF) built using Bi-Delta Networks. Functional equivalence between Bi-Delta Networks induces an equivalence between TBSFs as well. A weaker form of equivalence guaranteeing identical performance under conditions of symmetric traffic is studied, and a sufficient condition for weak equivalence between functionally distinct TBSFs is", } @INPROCEEDINGS{Stey93:General, AUTHOR="B. Steyart and H. Bruneel and Yongqiang Xiong", TITLE="A General Relationship between Buffer Occupancy and Delay in discrete-Time Multiserver Queueing Models, Applicable in {ATM} networks", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="1250-1258", MONTH="March/April", YEAR=1993, REFERENCES=20, KEYWORDS="ATM; queueing system; multiserver", ABSTRACT="In this paper, we investigate a multiserver discrete-time buffersystem. Packets arrive in the system according to a general correlated process, which is not further specified. The service times of the packets are of constant length. Explicit expressions are derived for the distribution, probability generating function, mean and variance of the packet delay, interms of the", } @INPROCEEDINGS{Taki93:Cell, AUTHOR="T. Takine and T. Suda and Toshiharu Hasegawa", TITLE="Cell Loss and Output Process Analyses of a Finite-Buffer discrete-Time {ATM} Queueing System with Correlated Arrivals", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=3, PAGES="1259-1269", MONTH="March/April", YEAR=1993, REFERENCES=20, KEYWORDS="ATM; statistical multiplexing; performance evaluation; analysis; DMAP; batch arrival; matrix analytic method; loss probability; Interdeparture Time; Busy Period; Output Process", ABSTRACT="This paper analyzes the performance of an ATM switching node considering cell arrival correlation. An ATM switching node is modeled as a discrete-time finite-buffer queue. Cell arrivals are assumed to follow a semi-Markovian process, where the numberof cell arrivals in a slot depends on the states of the underlying MC in the current and the previous slots. Obtained results include the cell loss probability, the consecutive loss", } @INPROCEEDINGS{Tedi93:Effectiveness, AUTHOR="T. E. Tedijanto and L. GŸn", TITLE="Effectiveness of Dynamic Bandwidth Management Mechanisms in {ATM} networks", BOOKTITLE=infocom, ADDRESS="San Francisco", VOLUME=1, PAGES="358-367", MONTH="March/April", YEAR=1993, REFERENCES=7, KEYWORDS="ATM; Congestion Control; Flow Control; Access Control; Policing Function; Traffic Measurement", ABSTRACT="One of the major chal