Reuven Cohen, Technion, Israel
Dan Pitt, Nortel, USA
Full Day Tutorials:
|
Tutorial 1 |
Wavelength-Routing Optical Networks |
Kumar Sivarajan Indian Institute of Science |
Monday,
9:00am - 5:00pm |
|
Tutorial 2 |
The Evolution of QoS in the Internet Standards Community |
Jon Crowcroft University College London |
Sunday,
9:00am - 5:00pm |
|
Tutorial 3 |
Overview of Network Security |
Radia Perlman Sun Microsystems |
Sunday,
9:00am - 5:00pm |
|
Tutorial 4 |
Teletraffic Models and Tools: From Basics to Advanced |
Khosrow Sohraby Univ. of Missouri, Kansas City |
Monday,
9:00am - 5:00pm |
|
Tutorial 5 |
IP Multicast: Past, Present and Future |
Radia Perlman, Sun Microsystems Christophe Diot, Sprint |
Monday,
9:00am - 5:00pm |
Half Day Tutorials:
|
Tutorial 6 |
MPLS |
Loa Andersson Nortel networks |
Sunday,
8:30am - 12:30pm |
|
Tutorial 7 |
New Technologies for LAN Systems |
Dono Van-Mierop IBM Haifa |
Monday,
8:30am - 12:30pm |
|
Tutorial 8 |
Satellite IP Networking |
Catherine Rosenberg Purdue University |
Monday,
1:30pm - 5:30pm |
|
Tutorial 9 |
Mobile IP: Adding Mobility to the Internet |
Charles Perkins Nokia Research |
Sunday,
1:30pm - 5:30pm |
Tutorial 1: Wavelength-Routing Optical Networks
Kumar Sivarajan,
Department of Electrical and Communication Engineering,
Indian Institute of Science
Second-generation optical networks can be conveniently thought of as constituting an *optical layer* above which lie conventional or first-generation networks such as SONET and ATM. In this tutorial we will first describe the optical layer and the services provided by it. We will then describe the motivations for the deployment of wavelength-routing networks as an optical layer. We will discuss the available architectural options and the tradeoffs involved. We will also discuss the effects of full/limited/no wavelength conversion on the design of these networks.
Two important problems that arise in the design of these networks are those of (a) routing and wavelength assignment (RWA), and (b) topology design. The RWA problem has many variations depending on the nature of the network, the traffic model assumed, and the performance measures of interest. We will discuss the RWA problem in detail, and make an up-to-date survey of the research results obtained in this area. We will suggest other approaches to the RWA problem which could be of interest of researchers.
Different services could be offered by a second generation optical network to the layer/network above it. One such service is the *lightpath* service where the network provides a dedicated optical path between two nodes (quite similar to a circuit-switched connection). We will study the topology design problem in the context of a wavelength-routing optical network offering such a lightpath service. The topology design problem now encompasses the design of two topologies: the physical or fiber topology of the wavelength-routing network, and the logical or lightpath topology which is the topology seen by a network that lies above the wavelength-routing network. The problem of routing lightpaths through the physical topology, and the problem of routing data traffic over the lightpath topology, may also have to be considered while designing the topologies. We will discuss the various issues that arise in this rather complex problem, and again provide an up-to-date survey of the work done in this area.
The provision of IP directly over WDM, rather than over another layer such as SONET which in turn uses the optical layer, has seen considerable interest in the last year. We will discuss the issues arising in the provision of IP over WDM and the tradeoffs involved.
Finally we will discuss the control and management of these networks, with particular emphasis on fault management techniques. Various available options for fault management will be discussed and the problems that remain to be tackled will be outlined.
BIOGRAPHY
Kumar N. Sivarajan has been an active researcher in the area of optical networks since 1990. He has coauthored a book, with Rajiv Ramaswami, titled "Optical Networks: A Practical Perspective" published by Morgan Kaufmann Publishers. He is the corecipient, with Rajiv Ramaswami, of the IEEE Communications Society 1996 William R. Bennett Prize Paper Award and the 1997 IEEE W. R. G. Baker Prize Award for the paper entitled "Routing and Wavelength Assignment in All-Optical Networks," published in the IEEE/ACM Transactions on Networking.
He has taught a graduate level course on optical networks for the past four years at the Indian Institute of Science. He was also conducted tutorials on optical networks at various conferences, most recently at GLOBECOM'98 and ECOC'99.
Tutorial 2: The Evolution of QoS in the Internet Standards Community
Jon Crowcroft,
Department of Computer Science, University College London
The Internet Engineering Task Force is the forum for the development of protocols for the Information Superhighway. Today, the Internet Protocol provides a best effort datagram service. End to end mechanisms and protocols provide for reliability and adapt traffic flows to network conditions to offer some semblance of a fair share service.
In lightly loaded networks, it is also possible to provide soft realtime applications with receiver adaption and thus offer one-to-one, one-to-many streaming, and many-to-many conferencing facilities.
However, the economics of capacity and demand often lead to large portions of the public Internet, and some Intranets being congested. In these circumstances, overloaded IP routers discard packets. While elastic applications based on TCP (one-to-one, reliable, bulk data transfer) will tolerate this, many other applications become either unusable, or uneconomic.
The IETF has gone through two phases of development of new protocols to try to solve this problem. Firstly, the Integrated Services (intserv) were defined to support specific Quality of Service targets, together with a signaling protocol, RSVP, to support online/on demand resource allocation. This definition phase was successful in bringing together experts in traffic management (often from other communities such as the IEEE and ATM Forum), but has failed in terms of delivering a suite that is deployable in the open Internet for a variety of reasons which will be discussed. The second phase of development has been led by gathering input users and ISPs, and is termed Differentiated Services (diffserv). These services are rooted in the idea of Class of Service and in the concepts of service level agreement. Another key problem solved by diffserv is the provision of flow and service aggregation.
In this tutorial we will look at this evolutionary process from an architectural standpoint. We will cover the rationale, the functional components of each approach, and their operational rules. The seminar will include some material on classical network design, sine this is usually assumed in the IETF models. Thus link placement and dimensioning, and admission control, metering, policing will be described. QoS mapping from the IP layer to lower levels (ATM, classical LANs, 802 p&q and so on). Novel problems that the IP community have solved will also be touched on - for example, fast packet classification, label switching, and lightweight approximations to WFQ implementation schemes will be explained and contrasted with QoS flow aggregation techniques.
BIOGRAPHY
Jon Crowcroft is a professor of networked systems in the Department of Computer Science, University College London, where he is responsible for a number of European and US funded research projects in Multi-media Communications. He has been working in these areas for over 18 years. He graduated in Physics from Trinity College, Cambridge University in 1979, and gained his MSc in Computing in 1981, and PhD in 1993. He is a member of the ACM, the British Computer Society and a Fellow of the IEE and a senior member of the IEEE. He is a member of the IAB and was general chair for the ACM SIGCOMM for the last 4 years. He is also on the editorial team for the ACM/IEEE Transactions on Networks. With Mark Handley, is the co-author of "WWW: Beneath the Surf" (UCL Press); he also authored "Open Distributed Systems" (UCL Press/Artech House). A third book, with Handley and Ian Waleman, "Internetworking Multimedia" (Taylor and Francis/Morgan Kaufman) was published in September 1999.
Tutorial 3: Overview of Network Security
Radia Perlman,
Sun Microsystems
Who are you, and should you be doing that? These are the basic questions that a system connected to a network should be pondering. Network security protocols help to answer these in a secure way. This workshop gives an overview of how such protocols work, including the basics of cryptography, key distribution, and protocol design pitfalls. It also covers the non-technical challenges facing designers of security systems, such as export laws and patent issues. Once we cover the basic techniques used by network security systems, we give an overview of many of the standards and commercially deployed systems.
BIOGRAPHY
Radia Perlman is a Distinguished Engineer at Sun Microsystems. She is known for her contributions to bridging (spanning tree algorithm) and routing (link state routing) and security (sabotage-proof networks). Radia is the author of "Interconnections: Bridges and Routers", and is co-author of Network Security: Private Communication in a Public World", both of which were listed in the top 10 most useful Networking reference books in the March, 1998 issue of Network Magazine. She was recently featured as one of the 25 people whose work has most influenced the industry in the 25th anniversary edition of Data Communications magazine. She has about 50 issued patents in the fields of routing and security. She has a PhD in computer science from MIT as well as S.B. and S.M. in mathematics from MIT.
Tutorial 4: Teletraffic Models and Tools: From Basics to Advanced
Khosrow Sohraby,
Dept. of Computer Networking, University of Missouri at Kansas City
In this tutorial, we provide a unified and simple treatment of teletraffic models and their teletraffic analysis in computer and telecommunications networks.
We first review the basic teletraffic models and their solution techniques. Then, in a systematic and simple manner, we extend the models to cover more practical and complex situations frequently encountered in modern networks.
A unified and powerful solution methodology based on elementary concepts in linear system theory will be presented. This will cover common teletraffic models including discrete state Markov models, fluid-flow models, Brownian motion models, and structured Markov Chains frequently encountered in the teletraffic analysis of modern networks. To further clarify the concepts, the usage of the publicly available software package TELPACK (TELetraffic PACKage) will be demonstrated by using applications in high-speed networks and cellular systems.
Outline:
It is recommended that students have some basic background of probability and Markov chains.
BIOGRAPHY
Khosrow Sohraby received his Ph.D degree in 1985 from the University of Toronto, Toronto, Canada, in Electrical Engineering. In 1986 he joined AT&T Bell Laboratories, Holmdel, NJ as a Member of Technical Staff in the Teletraffic Theory and System Performance Analysis Department. In 1989 he joined IBM T.J. Watson Research Center, Yorktown Heights, NY as a Research Staff Member in the Communications Systems Department. He joined the Computer Science Telecommunications division at the University of Missouri-Kansas City in 1994 as professor. Dr. Sohraby served as the technical program chair of IEEE INFOCOM '95 and the chair of 10th Annual IEEE Workshop on Computer Communications. and he was a consultant to Lucent Technologies, Bell Labs during the summer of 1996.
Tutorial 5: IP Multicast: Past, Present, and Future
Christophe Diot,
Sprint
Radia Perlman,
Sun Microsystems
This course will try to answer some of the questions than many researchers, network designers, or Internet users may have about multicast: What is multicast? What do we need it for? What are the possible applications? but also more deployment related questions such as Why are there so many versions? Is there a market for multicast? Is it deployable? Who is actually doing it? How do I manage it?
Answering these question will allow attendees to understand why multicast service is currently not widely available, and analyze the strengths and the problems of the current protocol architecture and service model.
This tutorial covers both the ISP perspective and the network protocol designer perspective. It discusses deployment experience with current protocols, both positive and negative.
Most of the day will consist of in-depth coverage of all the protocols involved in datagram multicast, current and future, including addressing, IGMP, DVMRP, PIM, CBT, MSDP, MOSPF, BGMP, MBGP, Simple Multicast, and Express. Emphasis will be on understanding the implications of the protocols, and the advantages and disadvantages, rather than on details like exact packet formats.
The tutorial will also discuss protocols for providing reliable multicast and congestion control, as well as security issues including key distribution to a large group with dynamic membership, and limiting joins and transmissions to authorized members.
It will also discuss areas of active research, such as congestion control.
Outline:
BIOGRAPHIES
Radia Perlman (see above)
Christophe Diot received a Ph.D. degree in Computer Science from INP Grenoble in 1991. He was a research scientist at INRIA Sophia Antipolis, working on new Internet architecture and protocols, for the last 5 years. Diot moved to Sprint Advanced Technology Laboratory in October 1998 to take the lead of the IP research group. His current interest is in deployable multicast and Internet resource control. Diot is also member of the ACM, member of the COST 264 action, and serves as an editor for ACM/IEEE Transactions on Networking. He is co-author of the chapter on multicast routing in Christian Huitema's book "Routing in the Internet" and he is co-chair of the SIGCOMM 2000 Program committee.
Tutorial 6: MPLS
Loa Andersson,
Nortel networks
The Multi Protocol Label Switching (MPLS) tutorial will give an overview of the technological background and requirements that lead to the development of the MPLS technologies and standards. The tutorial will explain some early proprietary proposals and their mapping and contributions to the MPLS standard. A detailed walk through of MPLS technologies will be given and the crucial design decisions explained. A set of concepts needed to understand to area will be given and the MPLS modeling will be explained. The controversies on how to solve some technological issues, e.g. loop prevention/detection, traffic engineering, explicit routing and MPLS VPNs will be highlighted. The relations between MPLS and different infrastructure technologies, e.g. Ethernet, PPP, ATM and FR, will be explained. MPLS will be compared to other solutions that address the same type of problems. The role of MPLS in Internetworking and the recent directions taken by the MPLS WG will be illustrated. Possible future potentials and trends for MPLS will be discussed.
BIOGRAPHY
Loa Andersson is director of the Routing Architecture Lab at the Nortel Networks Enterprise Solutions Technology Center, based in Stockholm (Sweden) and responsible for the EMEA operations of the Technology Center. Works currently with routing related topics and standardization or routing protocols. Active in standardization organizations like, e.g. IETF and ATM Forum, and one of the main contributors to the MPOA and MPLS standards. Has been active in the industries since early 80-ties and worked with telephony, telephone terminals and terminal adapters, access switches and group switches, ATM switches, including IP over ATM, and an application platform for telephony services.
Tutorial 7: New Technologies for LAN Systems
Dono Van-Mierop,
IBM Israel
LAN technologies continue to evolve. Link speeds have increased from 10Mb/s to 1000Mb/s in 20 years, matching Moore's law of doubling every 3 years. Another tenfold bandwidth increase is currently in the works and is likely to be ready in time to beat Moore?s law.
LANs have become very popular, and today most non-residential computers (over 100 million) are LAN attached. The rate of adding attachments keeps growing. The retention of the Ethernet frame format over this period of time has enabled the continued support of all existing and emerging LAN applications, in spite of the dramatic changes in the underlying technologies.
The main source for new LAN technologies is the IEEE 802 standards committee, which is responsive to new user requirements, and applies new technologies while assuring backward compatibility. The committee keeps introducing new link technologies (e.g., Ethernet at various speeds) as well as system level technologies (e.g., switching).
This tutorial introduces the new LAN technologies in their historic perspective and in the context of building complete LAN systems.
It starts with the basic Gigabit Ethernet link as defined in 802.3z for fiber and in 802.3ac for UTP. The fact that the tenfold leap to 1Gb/s was made only recently did not discourage the committee to start working on 10Gb/s. The tutorial covers the impact of new basic technologies such as SiGe, WDM and EDFA on recent proposals for 10Gb/s Ethernet.
Users who prefer linear speed increase can resort to the recently completed 802.3ad Link Aggregation standard. The tutorial rounds off the link properties with the 802.3x flow control mechanism that is used to prevent congestion and limit the need for buffering.
Next the tutorial presents LAN topology considerations and cable specifications based on EIA-556 structured wiring. Scaling LANs to large sizes was made possible by the 802.1D bridging standard, and the transition from shared to switched full duplex links. LAN switches enabled the construction of very high performance networks very economically. Additional efficiency may be gained with VLANs that are specified in the 802.1Q standard.
Several mechanism simplify the management of LAN systems. In particular the self managing feature of auto negotiation, the Spanning Tree protocol and GVRP. The tutorial presents the new proposals for Spanning Tree per VLAN and fast converging Spanning Tree. It also discusses LAN system management as specified in Clause 30 and in a number of IETF MIBs.
Quality of service and security are handled by the 802.1Q priority and VLAN mechanisms. The tutorial presents the mapping of higher layer constructs onto 802.1Q. It seems that the role of LAN technology is becoming more important in the context of global networks. The tutorial concludes with discussing the potential convergence of LANs with SANs, MANs and WANs.
BIOGRAPHY
Dono van-Mierop is a member of the IEEE 802.1 and 802.3 committees. He got his Ph.D. in Computer Science from UCLA in 1979, and his B.Sc. and M.Sc. from the EE department at the Technion, in 1970 and 1972, respectively. He is Currently a senior staff member in the area of communications at the IBM Haifa Research Lab. From 1995 to 1999 he worked on Gigabit Ethernet and Multi-layer Switching as VP Engineering and CTO at LanOptics. From 1989 to 1995 he managed Engineering and made technical contributions in the area of FDDI, Ethernet, Network Management and ATM LAN Emulation, and participated in key acquisitions as CTO staff at 3Com. From 1985 to 1989 he established and managed the R&D group at Fibronics, pioneering FDDI and other networking technologies. From 1980 to 1985 he was the Chief Architect at Intel Israel, developing the world's first Ethernet controller chip. Prior to that he worked as a researcher at USC/Information Sciences Institute.
Tutorial 8: Satellite IP Networking
Catherine Rosenberg,
School of Electrical & Computer Engineering, Purdue University
Satellite networks play an ever-increasing role in the public and private internets, due mostly to their large geographic coverage, their inherent broadcast capabilities as well as their fast deployment. Next Generation Satellite Systems (NGSS), in particular, will provide direct-to-user, two-way connectivity and thus represent an attractive solution for the provision of high-valued services such as high speed Internet access or virtual private network (Intranet) solutions.
This tutorial will address various networking issues related to NGSS. These systems will be characterized by the support of:
The tutorial will present NGSS and their characteristics and applications. A classification of these systems will be given and existing or proposed NGSS will be discussed in the context of this classification. In particular, the tutorial will describe geostationary (GEO) systems with no on-board processing, GEO systems with OBP, and Low Earth Orbit (LEO) systems with OBP.
The tutorial will then present the technical challenges and risk areas, in particular the following topics will be discussed in some detail: architectures, resource management, on-board processing, terminal as well as protocol and service adaptation.
The reminder of the tutorial will focus on IP and discuss problems such as TCP/IP and IP multicast over NGSS.
BIOGRAPHY
Catherine Rosenberg graduated from ENST-Bretagne, France in 1983, obtained the M.S. in Computer Science from UCLA, in 1984 and the Doctorat en Sciences from Universite de Paris XI, France in 1986. From 1987-1988 she was a Member of Technical Staff at AT&T Bell Laboratories. During 1988-1996, she was Associate Professor at the Department of Electrical and Computer Engineering, Ecole Polytechnique of Montreal. She was with Nortel Networks, UK from 1996 to 1999 where she created and headed the R&D Department in Broadband Satellite Networking. She has joined the faculty of Purdue University in August 1999 as an Associate Professor in the School of Electrical and Computer Engineering. Catherine Rosenberg is an Associate Editor for IEEE Communications Magazine, IEEE Communications Surveys and Telecommunications Systems. She has been and is involved in many conferences including IEEE INFOCOM, International Teletraffic Congress (ITC) and IFIP Broadband Communication. Her research interests are in Broadband Satellite Networks; in Traffic Engineering including QoS, Charging, Network Design; and in IP Mobility. She has authored over 50 papers in these areas and has filed several patents. ~
Tutorial 9: Mobile IP: Adding Mobility to the Internet
Charles Perkins,
Nokia Research
The Internet is growing by leaps and bounds, and likewise mobile computers are becoming more and more popular. When mobile computers move, and attach themselves to new networks within the Internet, they can use mobile-IP as a means to achieve seamless roaming transparently to application software. In this situation, transparent means that the applications work just as before, and don't need to be recompiled or reconfigured. Seamless means that roaming from one place to another occurs without inconvenience to the user. As long as a physical path exists for communication, the user might not even be aware when a cell boundary has been crossed. The objective of the seminar is to lay out all the necessary protocol technology to allow mobile computers to use mobile-IP, and to describe the relevant operation of other protocols which can be used to aid mobility.
In this tutorial, I will explore in detail all aspects of mobile-IP and other standard protocols that further simplify the operation of mobile computers in the Internet, including:
The seminar is intended for anyone who is interested in learning about how to use mobile-IP, create a home network for mobile users within their organization, or explore new Internet protocols and mobile computing. This includes programmers, administrators, network managers, and mobile computer users who are already familiar with using the Internet.
BIOGRAPHY
Charles E. Perkins is a Research Fellow at Nokia Research Laboratories, investigating mobile wireless networking and dynamic configuration protocols. He is the editor for several ACM and IEEE journals for areas related to wireless networking. He is serving as document editor for the mobile-IP working group of the Internet Engineering Task Force (IETF), and is author or co-author of standards-track documents in the mobileip, svrloc, dhc (Dynamic Host Configuration) and IPng working groups. Charles is also associate editor for Mobile Communications and Computing Review, the official publication of ACM SIGMOBILE, and is on the editorial staff for IEEE Internet Computing magazine. Charles has authored a book on Mobile IP, and has published a number of papers and award winning articles in the areas of mobile networking, ad-hoc networking, route optimization for mobile networking, resource discovery, and automatic configuration for mobile computers. Charles has served on the Internet Architecture Board (IAB) of the IETF, on various committees for the National Research Council, and is currently the chairperson of the Nomadicity Working Team of the Cross-Industry Working Team (XIWT).
Last updated 10/8/99