THIRD INTERNATIONAL SYMPOSIUM ON HIGH PERFORMANCE DISTRIBUTED COMPUTING (HPDC-3) Westin St. Francis Hotel, San Francisco, California August 2-5, 1994 SPONSORS: - IEEE Computer Society - Northeast Parallel Architectures Center at Syracuse University IN COOPERATION WITH: - ACM SIGCOMM - Rome Laboratory The International Symposium on High Performance Distributed Computing provides a forum for presenting the latest research findings that unify parallel and distributed computing. In HPDC environments, parallel or distributed computing techniques are applied to the solution of computationally intensive applications across networks of computers. This symposium follows two successful earlier conferences held in Syracuse, NY and Spokane, WA in 1992 and 1993, respectively. ======================= Pre-Symposium Tutorials ======================= Tuesday, August 2 8:30 AM -- 12 NOON Concurrent Sessions Tutorial 1A: Software Systems and Tools for High Performance Distributed Computing Anand Tripathi, University of Minnesota Tutorial 1B: Interfacing to High Speed Networks: Adapter Design and Operating System Issues. K. K. Ramakrishnan, Digital Equipment Corporation 2:00 PM -- 5:30 PM Concurrent Sessions Tutorial 2A: Message Passing Using MPI: from Fundamentals to Applications David W. Walker, Oak Ridge National Laboratory Tutorial 2B: High Performance Distributed Computing in a Supercomputing Environment: Computational Services and Applications Issues W. T. C. Kramer and Horst D. Simon, NASA Ames Research Center ====================================== Wednesday, August 3 ====================================== 8:00 AM - 10:00 AM Registration 8:30 AM - 10:00 AM Plenary Session Keynote Speech Robert E. Kahn, Corporation for Networking Research Initiatives 10:30 AM - 12:00 NOON SESSION 1: INVITED PAPERS Chair: W. Johnston, NYNEX Multimedia Supercomputing: The Use of Supercomputers to Drive High-Performance Multimedia Systems and Virtual Environments Rick Stevens, Argonne National Laboratory Constructing Numerical Software Libraries for HPCC Environments Jack Dongarra, University of Tennessee and Oak Ridge National Laboratory Overview of RWC Massively Parallel Computer Project Shuichi Sakai, Real World Computing Project, Japan 12:00 NOON - 1:30 PM LUNCH sponsored by Kendall-Square Research Speaker: TBA 1:30 PM - 3:00 PM Concurrent Sessions SESSION 2A: SOFTWARE TOOLS AND ENVIRONMENTS I Chair: M. Annaratone, Digital "The Virtual Computing Environment" P. Rouselle, P. Tymann, G. Fox, S. Hariri, Syracuse University "The Use of Frameworks for Scientific Computation in a Parallel Distributed Environment" R. Armstrong, Sandia National Laboratories J. Macfarlane, Lawrence Berkeley Laboratory "Falcon -- Toward Interactive Parallel Programs: The On-line Steering of a Molecular Dynamics Application" G. Eisenhauer, W. Gu, K. Schwan, N. Mallavarupu, Georgia Institute of Technology SESSION 2B: HIGH-SPEED NETWORKS AND APPLICATIONS Chair: W. Dabbous, INRIA, France "High-Performance TCP/IP and UDP/IP Networking in DEC OSF/1 Alpha AXP" C-H. Chang, D. Flower, J. Forecast, H. Gray, B. Hawe, A. Nadkarni, K. K. Ramakrishnan, U. Shikarpur, D. Ting, K. Wilde, Digital Equipment Corporation "Design and Implementation of Global Reduction Operations across ATM Networks" C. Huang, P. K. McKinley, Michigan State University "Traffic Monitoring for Capacity Allocation of Multimedia Traffic in ATM Broadband Networks" A. Burrell, D. Makrakis, P. Papantoni-Kazakos, University of Ottawa 3:30 PM - 4:30 PM Concurrent Sessions SESSION 3A: SOFTWARE TOOLS AND ENVIRONMENTS II Chair: A. Skjellum, Mississippi State University "WAVE Processing of Networks and Distributed Simulation" P. M. Borst, University of Karlsruhe, Germany M. Corbin, DRA, Farnborough, U.K. P. S. Sapaty, University of Surrey, U.K. "Providing High Performance Distributed Computing through Scalable Computation Servers" O. Kremien, Bar-Ilan University, Israel J. Kramer, Imperial College, U.K. SESSION 3B: HPDC APPLICATIONS I Chair: D. McAuliffe, Rome Laboratory "Running a Climate Model in a Heterogeneous Distributed Computer Environment" C. R. Mechoso, D. J. Farrara, J. A. Spahr, University of California, Los Angeles "Distributed Computation of Electromagnetic Scattering Problems Using Finite-Difference Time-Domain Decompositions" Sandy Nguyen, Brian Zook, Southwest Research Institute Xiaodong Zhang, The University of Texas at San Antonio ====================================== Thursday, August 4 ====================================== 9:00 AM - 10:00 AM Keynote Speech: System Architectures for High Performance Computing in the 1990's Tilak Agerwala, IBM 10:30 AM - 12:00 NOON Concurrent Sessions SESSION 4A: MAPPING AND SCHEDULING I Chair: A. Grimshaw, University of Virginia "Scheduling Large-Scale Parallel Computations on Networks of Workstations" R. D. Blumofe, D. S. Park, Massachusetts Institute of Technology "Scheduling a Metacomputer by an Implicit Voting System" K. Kremer, University of Technology Aachen, Germany F. Ramme, University of Paderborn, Germany "A Decomposition Advisory System for Heterogeneous Data-Parallel Processing" P. E. Crandall, M. J. Quinn, Oregon State University SESSION 4B: DISTRIBUTED SHARED-MEMORY SYSTEMS Chair: T. V. Lakshman, Bellcore "Evaluating Weak Memories with Maya" D. Agrawal, M. Choy, H. Va Leong, A. Singh University of California, Santa Barbara "Analysis and Transformation of Parallel Programs for Fast Data Sharing" A. Li, University of Victoria, Canada G Hermannsson, L. Wittie, State University of New York, Stony Brook "An Experimental Active-Memory-Based Network Environment" A. Asthana, M. Cravatts, P. Krzyzanowski, AT&T Bell Laboratories 12:00 NOON - 1:30 PM LUNCH 1:30 PM - 3:00 PM Panel GOVERNMENT STRATEGIES FOR HIGH-PERFORMANCE DISTRIBUTED COMPUTING Panel Chair: J. A. Graniero, Rome Laboratory Panelists: Hank Dardy, Naval Research Laboratory Andrew White, Los Alamos Richard Freund, Naval Research and Development Center L/C John Toole, ARPA/CSTO L/C Larry Davis, AFOSR/ST Ray Kline, Sandia National Laboratories 3:30 PM - 5:00 PM Concurrent Sessions SESSION 5A: PARTITIONING AND LOAD BALANCING Chair: R. Perrot, Queens University, Belfast "Network Partitioning of Data Parallel Programs" J. B. Weissman, A. S. Grimshaw, University of Virginia "Exploiting Inter-Task Dependencies for Dynamic Load Balancing" W. Becker, G. Waldmann, University of Stuttgart, Germany "Automatic Generation of Parallel Programs with Dynamic Load Balancing" B. S. Siegell, P. Steenkiste, Carnegie Mellon University SESSION 5B: HPDC APPLICATIONS II Chair: J. Patterson, Boeing "Distributed Solutions to the Delay Fault Test Quality Evaluation Problem" I. Pramanick, A. K. Pramanick, IBM "Design and Implementation of Parallel Algorithms for Gene Finding" J. J. Puthukattukran, S. Chalasani, University of Wisconsin-Madison P. Senapathy, Genome International Corporation "Solving Partial Differential Equations on a Network of Workstations" C.-C. Hui, M. Hamdi, Ishfaq Ahmad, Hong Kong University of Science and Technology ====================================== Friday, August 5 ====================================== 8:30 AM - 10:00 AM Concurrent Sessions SESSION 6A: MAPPING AND SCHEDULING II Chair: Ardsher Ahmed, UMASS Dartmouth Parallel Computations on the CHARM Heterogeneous Workstation Cluster" V. A. Saletore, J. Jacob, M. Padala, Oregon State University "Mapping Parallel Iterative Algorithms onto Workstations Networks" A. Heddaya, K. Park, Boston University "A Task Graph Centroid" C. Leangsuksun, J. Potter Kent State University SESSION 6B: FAULT-TOLERANCE AND I/O IN DISTRIBUTED SYSTEMS Chair: I. Ahmad, Hong Kong University of Science & Technology "Data Reshuffling in Support of Fast I/O for Distributed-Memory Machines" C. Bornstein, P. Steenkiste Carnegie Mellon University "Process-Replication Technique for Fault-Tolerance and Performance Improvement in Distributed Computing Systems" J-F. Chiu, G-M. Chiu, National Taiwan Institute of Technology "Performance Evaluation of a Partial Dynamic Declustering Disk Array System" V. Catania, A. Puliafito, S. Riccobene, L. Vita, University of Catania, Italy 10:30 AM - 12:00 NOON Concurrent Sessions SESSION 7A: NETWORK PROTOCOLS AND INTERFACES Chair: I. Akyildiz, Georgia Institute of Technology "Large-Scale Group Communication Protocol on High-Speed Channel" M. Takamura, M. Takizawa, Tokyo Denki University "Deciding Boundedness for Systems of Communicating Finite State Machines" A. Benslimane, Universite de Franche-Comte, France "Design of the Header Processor for the PSi Implementation of the Logical Link Control Protocol in LANs" F. A. Morales, H. Abu-Amara, Texas A&M University SESSION 7B: PERFORMANCE EVALUATION AND RESOURCE MANAGEMENT Chair: J. J. Garcia-Luna, University of California, Santa Cruz "PMT: A Tool to Monitor Performance in Distributed Systems" V. Catania, O. Granato, A. Puliafito, L. Vita, University of Catania, Italy "Reducing Variations in Parallel Efficiency for Unstructured Grid Computations" B. Worner, G. Geunder, M. Hardtner, R. Zink, University of Stuttgart, Germany "Performance Prediction for Different Consistency Schemes in Distributed Shared Memory Systems" Z. G. Vranesic, S. Srbljic, University of Toronto L. Budin, University of Zagreb, Croatia ______________________________________________________________ KEYNOTE SPEAKERS Robert E. Kahn is President of the Corporation for National Research Initiatives (CNRI), which he founded in 1986 after a 13- year term at the Advanced Research Projects Agency (ARPA). Dr. Kahn holds a Ph.D. from Princeton University. Prior to his service at CNRI he held positions at Bell Laboratories, MIT and DARPA where he became Director of DARPA's Information Processing Techniques Office (IPTO) and initiated the United States government's billion dollar Strategic Computing Program, the largest computer research and development program ever undertaken by the federal government. Dr. Kahn is a member of the National Academy of Engineering, a Fellow of the IEEE, a Fellow of AAAI, a recipient of the AFIPS Harry Goode Memorial Award, the Marconi Award, the ACM SIGCOMM Award, the President's Award from the ACM, the IEEE Koji Kobayashi Computer and Communications Award, the ACM Software Systems Award, and the ASIS Special Award. He was twice the recipient of the Secretary of Defense Meritorious Civilian Service Award. Tilak Agerwala is Director of Architecture and System Design for IBM's parallel computer products. He is responsible for future product designs, performance analysis, and technical strategy. Dr. Agerwala has held executive positions in the RISC System 6000 division, where he was responsible for future system technology development, and at the T.J. Watson Research Center, where he initiated and directed broad research programs in parallel processing, artificial intelligence, and supercomputing. Dr. Agerwala is a member of the IBM Academy of Technology and has served on the Corporate Technical Committee. He was elected a Fellow of the IEEE for his leadership in the development of very high performance computers. ********************************************************************** DESCRIPTIONS OF TUTORIALS --------------------------------------------------------------------------- Tutorial 1A Software Systems and Tools for High Performance Distributed Computing Anand Tripathi, University of Minnesota The objective of this tutorial is to present an overview of the mechanisms and software tools/systems that are currently available for high-performance distributed programming in local-area networks. A high performance distributed programming environment needs mechanisms for specifying and managing parallel computation structures of a distributed application, transparent distribution of load in the network, dynamic linking of components, error detection and recovery, and transparency of the hardware/software heterogeneity in the network. This tutorial will present an overview of the most commonly used paradigms and models for distributed computing. These are related to interprocess communication models such as message passing and the remote procedure call (RPC), heterogeneous computing, load balancing and scheduling of tasks, and naming and protection of resources in the network. Reliability related issues in implementing RPC will be discussed. An overview the object model of computing will be presented in the context of micro-kernel architectures for distributed computing. The concept of group is a useful abstraction for managing a collection of related activities. Primitives for group management and broadcast communication will be presented. A number of software tools and programming languages are currently available for high performance distributed programming. An overview of the salient features and capabilities of Parallel Virtual Machine (PVM), Linda, Express, P4, CODE/ROPE, Mentat, CC++, PCN, Jade, Orca, Concurrent C, SR will be presented. SPEAKER: Anand Tripathi is an Associate Professor in the Department of Computer Science, University of Minnesota, Minneapolis. He received the B.Tech degree from the Indian Institute of Technology, Bombay, India, 1972, and the M.S. and Ph.D. degrees >from the University of Texas at Austin in 1978 and 1980, respectively, in Electrical Engineering. From 1972 to 1975 he worked as a research scientist at Bhabha Atomic Research Center, Bombay, India. During 1981 to 1984 he worked as a Senior Principal Research Scientist at Honeywell Computer Sciences Center, Bloomington, MN. At the University of Minnesota he has led the design and development of the Nexus distributed operating system and its programming environment. His research interests include parallel and distributed systems, object-oriented programming, and fault-tolerant computing. In the past he has presented tutorials on distributed computing systems at some of the IEEE sponsored conferences and tutorial-weeks. --------------------------------------------------------------------------- Tutorial 1B Interfacing to High Speed Networks: Adapter Design and Operating System Issues K. K. Ramakrishnan, Digital Equipment Corporation Overview: We have seen significant increases in the bandwidth available for computer communication networks in the recent past. Commercially available Local Area Networks operate at 100 Mbits/sec and research networks are running at greater than 1 Gbit/sec. Host CPU processing speed has also increased relentlessly. However, the anticipation that end-user applications can effectively use these large communication link speeds has yet to be fulfilled to a large extent. Network I/O at the end-system has often been perceived as the bottleneck for distributed applications. This tutorial discusses the important topics in the design of a high-speed network adapter, including the memory-system architecture, buffer management, support for Quality of Service (QoS) guarantees and flow/congestion control, and host software issues. Several case studies will be presented. Speaker: K. K. Ramakrishnan is a Consulting Engineer in the Distributed Systems Architecture and Performance group at Digital Equipment Corporation. He has been with Digital since 1983 after completing his Ph.D. in Computer Science from the University of Maryland. His group is involved in architecting Digital's efforts in the networking and distributed systems area. Dr. Ramakrishnan has worked and published papers in the areas of load balancing, congestion control and avoidance, algorithms for FDDI, distributed systems performance and issues relating to network I/O. Dr. Ramakrishnan participates in the Internet Engineering Task Force and is a member of the End-End Group, as part of the Internet Research Task Force. He is also a technical editor for IEEE Network. --------------------------------------------------------------------------- Tutorial 2A Message Passing Using MPI: from Fundamentals to Applications David W. Walker, Oak Ridge National Laboratory This tutorial will describe the features of the MPI message passing standard, and will show how to use MPI in applications. The tutorial is intended to benefit researchers who have some experience with message passing, and who wish to assess the advantages of MPI for their particular applications. However, the tutorial will be structured to also be of use to novices. The tutorial will be divided into three main parts. The first part will give an overview of MPI, describe how it came about, and will discuss the basic point-to-point and collective communication capabilities of MPI. The second part will describe advanced features in MPI, in particular, the management of groups and communication contexts. The third part will be devoted to the presentation of application kernels and examples written using MPI. SPEAKER: David Walker is a Research Staff Member in the Mathematical Sciences Section at Oak Ridge National Laboratory, and an Adjunct Associate Professor in the Department of Computer Science of the University of Tennessee, Knoxville. He obtained his B.A. in Mathematics from Jesus College, Cambridge, in 1973, his M.S. in Astrophysics in 1979, and his Ph.D. in Physics from the University of London in 1983. From 1986-88. Dr. Walker was a member of the research staff of the Concurrent Computation Project at the California Institute of Technology, and from 1988-1990 was an Associate Professor in the Department of Mathematics at the University of South Carolina. He was worked at ORNL since 1990, where he is mainly involved in the design of software libraries, algorithms, and application for distributed memory concurrent computers. Dr. Walker was instrumental in establishing the MPI Forum, which led to the specification of the MPI standard and in which he played an active role. --------------------------------------------------------------------------- Tutorial 2B High Performance Distributed Computing in a Supercomputing Environment: Computational Services and Applications Issues W. T. C. Kramer and Horst D. Simon, NASA Ames Research Center The focus of this tutorial will be a discussion of current hardware and software trends for massively parallel supercomputers >from the perspective of applications. In case studies the lessons learned at NASA Ames will be presented. The main thrust of the tutorial will be a presentation of high performance computing topics which will remain relevant for a long period of time, independent of currently ``hot" machines. This requires a detailed investigation of the issue of communication. From an applications perspective the communication characteristics of various applications will be examined. A new taxonomy of parallel application will be developed. Similarly high-performance architectures will be investigated with respect to their communications performance. The matching of the applications taxonomy with the architectural characteristics of a machine will form the basis for the understanding of high performance computing. High performance distributed computing will thus be evaluated as one possible alternative to both MPP and PVP systems. Emphasis will be placed on an honest evaluation of the capabilities and challenges that distributed computing faces in a production environment. The tutorial utilizes the unique experience made at the NAS facility at NASA Ames Research Center. Over the last five years at NAS massively parallel supercomputers such as the Connection Machines CM-2 and CM-5 from Thinking Machines Corporation and the iPSC/860 (Touchstone Gamma Machine) and Paragon Machines >from Intel were used in a production supercomputer center alongside traditional vector supercomputers such as the Cray Y-MP and C90. In addition to these resources, NAS has operated since 1993 a distributed computing testbed (DCT), consisting of an environment with 50 SGI and 40 SUN workstations. Speakers: William Kramer is Chief of the NAS Computational Service Branch, and is responsible for providing state-of-the-art support and enhancement of a Cray YMP, two C-90s, CM-5, an Intel iPSC/860, an Intel Paragon and several Convex systems, along with over 250 other systems and workstations connected via Ultranet, FDDI and ethernet. These systems are used by a nationwide network of over 1400 researchers in 100+ sites, connected with the advanced NAS Aeronet and other networks. Bill was responsible for the first production Cray-2 and YMP at NAS before becoming Branch Chief. He is currently leading the NAS effort in computing on loosely clustered workstations. He has authored or presented papers sessions and seminars on a number of subjects, including system management, computer graphics, security and supporting advanced users at supercomputer centers. He and his staff presented the highest rated Tutorial at Supercomputing 89, "System Management in the UNIX Supercomputing Environment", which was presented again by invitation at SC '90. Horst D. Simon is with Computer Sciences Corporation at the Applied Research Branch at the NAS (National Aerodynamics Simulator) Systems Division at NASA Ames Research Center in Moffett Field, California. He is CSC department manager and responsible for a group of researchers in the areas of parallel algorithm development and scientific visualization, who as contractor personnel collaborate with the NAS staff. His research interests are in the development of high performance algorithms for vector and parallel machines. Particular areas of interest are sparse matrix algorithms, algorithms for large scale eigenvalue problems, and domain decomposition algorithms for irregular domains for parallel processing. Simon's algorithm research efforts were honored with the 1988 Gordon Bell Award for parallel processing research. Previously, until October 1989, Dr. Simon was an employee of Boeing Computer Services, and >from October 1987 through October 1989, supported the NAS project in a similar capacity. >From 1982 to 1983 Dr. Simon was an assistant professor at the Department of Applied Mathematics, SUNY Stony Brook, New York. Dr. Simon holds a Ph.D. in mathematics from the University of California (1982), Berkeley and a Diplom in Mathematik from the TU Berlin, Germany (1978). Dr. Simon has presented tutorials on high performance computing at all previous Supercomputing conferences. His tutorials have consistently been rated excellent or very good. --------------------------------------------------------------------------- CONFERENCE LOCATION HPDC-3 will be held at the Westin St. Francis Hotel in downtown San Francisco, California. Just 30 minutes from the airport, the Westin St. Francis is in the heart of the city on Union Square surrounded by fine restaurants, shops and theaters. Cable cars stop at the front door. The hotel is within walking distance of San Francisco's Chinatown and an easy cable car ride away from Ghiradelli Square and Fisherman's Wharf. Napa and Sonoma Valleys, the heart of California wine country, are an hour's drive away. If you have additional time before or after the symposium you may want to explore Marin County's redwoods, the California Mission, or drive scenic highway 101 to San Simeon. Monterey, Carmel, Santa Cruz, Yosemite National Park and Lake Tahoe are not far away. Weather: San Francisco in August is typically very pleasant. Sunny and warm (about 70F) during the day, and cool (about 50F) in the evenings. Note that the evening fog often makes it feel a bit colder and a lightweight jacket or all-weather coat is advised. Transportation: The hotel is about 20 minutes away from San Francisco International Airport. A taxi ride from the airport typically costs about $25. Also, the SFO Airporter has direct service to the hotel at 20-minute intervals. The fare is $10 one-way and $17 round-trip. Rental Cars: HERTZ has been appointed as the car rental supplier for HPDC-3. Special rates with unlimited mileage have been offered to HPDC-3 attendees and are also available the week prior and the week after the symposium. To make reservations call HERTZ at 1-800-654-2240 (In Canada 1-800-263-0600) and identify yourself as HPDC attendee (Meeting # 15329). Cars can be rented at most Bay-area airports and also at the Westin St. Francis Hotel. Social Event: An evening dinner-cruise on the San Francisco Bay is planned for Wednesday, August 3. The cost of the cruise is included in the registration fee. Additional tickets may be purchased at a price of $55 each. E-mail service will be available at the hotel to the attendees. ************************************************************************** REGISTRATION INFORMATION Advance registration is available using the form below through July 8th. E-mail registration is available only through July 27th, and must use a credit card number. On-site registration at the Westin St. Francis Hotel will be available starting Tuesday, August 2nd, and every day of the conference starting at 8 AM. --------------------------------------------------------------------------- HPDC-3 Registration Form Please send this form and a check, credit card information, or money order (no purchase orders) to the address below. Make checks payable to SYRACUSE UNIVERSITY. Registrations accepted via postal mail, FAX, or email only. SU Conference Planning HPDC-3 Conference P.O. Box 4709 Syracuse, NY 13221-4709 Phone: (315) 443-3333 Fax: (315) 443-1168 e-mail: hpdc@nova.npac.syr.edu ----------------------------------------------------------- Symposium Registration (select one): Advance Registration Regular Registration (Received by July 8, 1994) (Received after July 8, 1994) --------------------- ------------------- IEEE/ACM SIGCOMM Member [ ] $325 [ ] $395 Non-Member [ ] $405 [ ] $485 Full-time student [ ] $180 [ ] $180 Symposium registration fee includes a copy of the proceedings, sponsored lunches, coffee breaks, and the Bay cruise. Student registration does not include proceedings or the Bay cruise. Extra copies of the proceedings may be purchased on-site. ----------------------------------------------------------- Extra Dinner/Cruise Tickets ___ @ $55 each ----------------------------------------------------------- Tutorials (check the box for each tutorial attending) Note: You may select at most one tutorial from the morning session and one from the afternoon session. _____ 1A: Software Systems and Tools for High Performance | | Distributed Computing...............................|___| _____ 1B: Interfacing to High Speed Networks..................| | |___| _____ 2A: Message Passing Using MPI...........................| | |___| _____ 2B: High Performance Distributed Computing | | in a Supercomputing Environment.....................|___| Enter total number of tutorials at the appropriate rate (rates are per tutorial). Advance Registration Regular Registration (Received by July 8, 1994) (Received after July 8, 1994) --------------------- ------------------- IEEE/ACM SIGCOMM Member ___ @ $165 ___ @ $195 Non-Member ___ @ $205 ___ @ $245 Total Tutorial Fee: $________US Each Tutorial registration fee includes attendance at one tutorial session, notes, and coffee breaks. There are no student fees for the tutorials. Cancellations of tutorial registrations after July 22 will be subject to the total fee. We reserve the right to cancel the tutorials due to insufficient participation or other unforeseeable problems, in which case fees will be refunded. ----------------------------------------------------------- Symposium Registration Fee $_________ + Extra cruise tickets $_______ + Tutorial Fees $___________ = Total Amount enclosed: $________ US ----------------------------------------------------------- Name: ____________________________________ Affiliation: _____________________________ Address: _________________________________ City/State/Zip/Postal Code/Country: ________________________________ Phone: ___________________________________ FAX: _____________________________________ E-mail Address: ___________________________ IEEE/ACM SIGCOMM Member #: ____________________________ (required for member rate) Method of Payment __ Visa __ Mastercard __check enclosed (payable to Syracuse University) Card Holder Name: ________________________ (as it appears on card) Card Number: _____________________________ Expiration Date: _________________________ Signature: _______________________________ You may register by e-mail using your credit card. Send completed registration to hpdc@nova.npac.syr.edu If you have a disability and may require accommodation in order ______ to fully participate in this activity, please check | | here. You will be contacted to discuss your needs. |____| Vegetarian Meals: ___ YES / ___ NO Cancellation Policy: All requests for refunds must be received in writing on or before July 22, 1994. No refunds will be made on cancellations made after this date. ************************************************************************** HOTEL REGISTRATION Please contact the Westin St. Francis Hotel directly for accommodations. The Westin St. Francis Union Square 335 Powell Street San Francisco, CA 94102 USA Phone: (415) 397-7000 Special conference rates have been arranged for attendees of HPDC-3. The rates are as follows: Type of Room Main Building Towers Single Double Single Double Standard $120 $120 - - Medium $135 $135 - - Deluxe $150 $150 $170 $170 Reservation cut-off date is Friday, July 1, 1994; Reservations made after this date will be on a space-available basis only. In order to receive our special rates you must identify yourself as a participant in the HPDC-3 Conference. ************************************************************************** CONFERENCE COMMITTEES ************************************************************************** SYMPOSIUM GENERAL CHAIR: Geoffrey Fox, NPAC, Syracuse University SYMPOSIUM STEERING COMMITTEE: - Salim Hariri, Syracuse University (Chair) - Tilak Agerwala, IBM - Andrew Grimshaw, University of Virginia - H. T. Kung, Harvard University - T. V. Lakshman, Bell Communications Research - Daniel McAuliffe, Rome Laboratory - C. S. Raghavendra, Washington State University PROGRAM CHAIR: Anujan Varma, University of California, Santa Cruz PUBLICITY CHAIRS: North America: T. V. Lakshman, Bell Communications Research Europe: Walid Dabbous, INRIA, France Asia: Makoto Takizawa, Tokyo Denki University, Japan TUTORIALS CHAIR: Ian F. Akyildiz, Georgia Tech EXHIBITS: C. S. Raghavendra, Washington State University REGISTRATION AND LOCAL ARRANGEMENTS: Peggy Van Arnam, Syracuse University PROGRAM COMMITTEE - Dharma Agrawal, North Carolina State University - Prathima Agrawal, AT&T Bell Labs - Ian F. Akyildiz, Georgia Tech - Marco Annaratone, DEC - Ken Birman, Cornell University - Suresh Chalasani, University of Wisconsin, Madison - Monsong Chen, IBM Research - Roger Chen, Syracuse University - Abdelaziz Chihoub, Siemens Corporate Research - Jon Crowcroft, University College London - Walid Dabbous, INRIA, France - Patrick Dowd, SUNY Buffalo - Dennis Duke, SCRI/Florida State University - Stuart Elby, NYNEX Science and Technology - Richard Freund, NRaD - J.J. Garcia-Luna, University of California, Santa Cruz - Arif Ghafoor, Purdue University - Andrew Grimshaw, University of Virginia - Salim Hariri, Syracuse University - S. H. Hosseini, University of Wisconsin, Milwaukee - T. V. Lakshman, Bell Communications Research - C. R. Mechoso, UCLA - Paul Messina, Caltech - Richard C. Metzger, Rome Laboratory - Paul Mockapetris, USC/ISI - John Morrison, Los Alamos National Laboratory - John Nicholas, Battelle Pacific Northwest Lab - James C. Patterson, Boeing Co. - Ira Pramanick, IBM - Michael Quinn, Oregon State University - C. S. Raghavendra, Washington State University - Paul Rupert, Lawrence Livermore National Laboratory - Karsten Schwan, Georgia Institute of Technology - Nita Sharma, Ncube Inc. - Vaidy Sunderam, Emory University - Makoto Takizawa, Tokyo Denki University, Japan - Alexander Thomasian, IBM Research - Satish Tripathi, University of Maryland - Larry Wittie, SUNY Stony Brook - Pen-Chung Yew, University of Illinois ************************************************************************** HPDC-3 CALL FOR PARTICIPATION ************************************************************************** THIRD INTERNATIONAL SYMPOSIUM ON HIGH PERFORMANCE DISTRIBUTED COMPUTING (HPDC-3) Westin St. Francis Hotel, San Francisco, California August 2-5, 1994 SPONSORS: - IEEE Computer Society - Northeast Parallel Architectures Center at Syracuse University IN COOPERATION WITH: - ACM SIGCOMM - Rome Laboratory The International Symposium on High Performance Distributed Computing provides a forum for presenting the latest research findings that unify parallel and distributed computing. In HPDC environments, parallel or distributed computing techniques are applied to the solution of computationally intensive applications across networks of computers. This symposium follows two successful earlier conferences held in Syracuse, NY and Spokane, WA in 1992 and 1993, respectively. ======================= Pre-Symposium Tutorials ======================= Tuesday, August 2 8:30 AM -- 12 NOON Concurrent Sessions Tutorial 1A: Software Systems and Tools for High Performance Distributed Computing Anand Tripathi, University of Minnesota Tutorial 1B: Interfacing to High Speed Networks: Adapter Design and Operating System Issues. K. K. Ramakrishnan, Digital Equipment Corporation 2:00 PM -- 5:30 PM Concurrent Sessions Tutorial 2A: Message Passing Using MPI: from Fundamentals to Applications David W. Walker, Oak Ridge National Laboratory Tutorial 2B: High Performance Distributed Computing in a Supercomputing Environment: Computational Services and Applications Issues W. T. C. Kramer and Horst D. Simon, NASA Ames Research Center ====================================== Wednesday, August 3 ====================================== 8:00 AM - 10:00 AM Registration 8:30 AM - 10:00 AM Plenary Session Keynote Speech Robert E. Kahn, Corporation for Networking Research Initiatives 10:30 AM - 12:00 NOON SESSION 1: INVITED PAPERS Chair: W. Johnston, NYNEX Multimedia Supercomputing: The Use of Supercomputers to Drive High-Performance Multimedia Systems and Virtual Environments Rick Stevens, Argonne National Laboratory Constructing Numerical Software Libraries for HPCC Environments Jack Dongarra, University of Tennessee and Oak Ridge National Laboratory Overview of RWC Massively Parallel Computer Project Shuichi Sakai, Real World Computing Project, Japan 12:00 NOON - 1:30 PM LUNCH sponsored by Kendall-Square Research Speaker: TBA 1:30 PM - 3:00 PM Concurrent Sessions SESSION 2A: SOFTWARE TOOLS AND ENVIRONMENTS I Chair: M. Annaratone, Digital "The Virtual Computing Environment" P. Rouselle, P. Tymann, G. Fox, S. Hariri, Syracuse University "The Use of Frameworks for Scientific Computation in a Parallel Distributed Environment" R. Armstrong, Sandia National Laboratories J. Macfarlane, Lawrence Berkeley Laboratory "Falcon -- Toward Interactive Parallel Programs: The On-line Steering of a Molecular Dynamics Application" G. Eisenhauer, W. Gu, K. Schwan, N. Mallavarupu, Georgia Institute of Technology SESSION 2B: HIGH-SPEED NETWORKS AND APPLICATIONS Chair: W. Dabbous, INRIA, France "High-Performance TCP/IP and UDP/IP Networking in DEC OSF/1 Alpha AXP" C-H. Chang, D. Flower, J. Forecast, H. Gray, B. Hawe, A. Nadkarni, K. K. Ramakrishnan, U. Shikarpur, D. Ting, K. Wilde, Digital Equipment Corporation "Design and Implementation of Global Reduction Operations across ATM Networks" C. Huang, P. K. McKinley, Michigan State University "Traffic Monitoring for Capacity Allocation of Multimedia Traffic in ATM Broadband Networks" A. Burrell, D. Makrakis, P. Papantoni-Kazakos, University of Ottawa 3:30 PM - 4:30 PM Concurrent Sessions SESSION 3A: SOFTWARE TOOLS AND ENVIRONMENTS II Chair: A. Skjellum, Mississippi State University "WAVE Processing of Networks and Distributed Simulation" P. M. Borst, University of Karlsruhe, Germany M. Corbin, DRA, Farnborough, U.K. P. S. Sapaty, University of Surrey, U.K. "Providing High Performance Distributed Computing through Scalable Computation Servers" O. Kremien, Bar-Ilan University, Israel J. Kramer, Imperial College, U.K. SESSION 3B: HPDC APPLICATIONS I Chair: D. McAuliffe, Rome Laboratory "Running a Climate Model in a Heterogeneous Distributed Computer Environment" C. R. Mechoso, D. J. Farrara, J. A. Spahr, University of California, Los Angeles "Distributed Computation of Electromagnetic Scattering Problems Using Finite-Difference Time-Domain Decompositions" Sandy Nguyen, Brian Zook, Southwest Research Institute Xiaodong Zhang, The University of Texas at San Antonio ====================================== Thursday, August 4 ====================================== 9:00 AM - 10:00 AM Keynote Speech: System Architectures for High Performance Computing in the 1990's Tilak Agerwala, IBM 10:30 AM - 12:00 NOON Concurrent Sessions SESSION 4A: MAPPING AND SCHEDULING I Chair: A. Grimshaw, University of Virginia "Scheduling Large-Scale Parallel Computations on Networks of Workstations" R. D. Blumofe, D. S. Park, Massachusetts Institute of Technology "Scheduling a Metacomputer by an Implicit Voting System" K. Kremer, University of Technology Aachen, Germany F. Ramme, University of Paderborn, Germany "A Decomposition Advisory System for Heterogeneous Data-Parallel Processing" P. E. Crandall, M. J. Quinn, Oregon State University SESSION 4B: DISTRIBUTED SHARED-MEMORY SYSTEMS Chair: T. V. Lakshman, Bellcore "Evaluating Weak Memories with Maya" D. Agrawal, M. Choy, H. Va Leong, A. Singh University of California, Santa Barbara "Analysis and Transformation of Parallel Programs for Fast Data Sharing" A. Li, University of Victoria, Canada G Hermannsson, L. Wittie, State University of New York, Stony Brook "An Experimental Active-Memory-Based Network Environment" A. Asthana, M. Cravatts, P. Krzyzanowski, AT&T Bell Laboratories 12:00 NOON - 1:30 PM LUNCH 1:30 PM - 3:00 PM Panel GOVERNMENT STRATEGIES FOR HIGH-PERFORMANCE DISTRIBUTED COMPUTING Panel Chair: J. A. Graniero, Rome Laboratory Panelists: Hank Dardy, Naval Research Laboratory Andrew White, Los Alamos Richard Freund, Naval Research and Development Center L/C John Toole, ARPA/CSTO L/C Larry Davis, AFOSR/ST Ray Kline, Sandia National Laboratories 3:30 PM - 5:00 PM Concurrent Sessions SESSION 5A: PARTITIONING AND LOAD BALANCING Chair: R. Perrot, Queens University, Belfast "Network Partitioning of Data Parallel Programs" J. B. Weissman, A. S. Grimshaw, University of Virginia "Exploiting Inter-Task Dependencies for Dynamic Load Balancing" W. Becker, G. Waldmann, University of Stuttgart, Germany "Automatic Generation of Parallel Programs with Dynamic Load Balancing" B. S. Siegell, P. Steenkiste, Carnegie Mellon University SESSION 5B: HPDC APPLICATIONS II Chair: J. Patterson, Boeing "Distributed Solutions to the Delay Fault Test Quality Evaluation Problem" I. Pramanick, A. K. Pramanick, IBM "Design and Implementation of Parallel Algorithms for Gene Finding" J. J. Puthukattukran, S. Chalasani, University of Wisconsin-Madison P. Senapathy, Genome International Corporation "Solving Partial Differential Equations on a Network of Workstations" C.-C. Hui, M. Hamdi, Ishfaq Ahmad, Hong Kong University of Science and Technology ====================================== Friday, August 5 ====================================== 8:30 AM - 10:00 AM Concurrent Sessions SESSION 6A: MAPPING AND SCHEDULING II Chair: Ardsher Ahmed, UMASS Dartmouth Parallel Computations on the CHARM Heterogeneous Workstation Cluster" V. A. Saletore, J. Jacob, M. Padala, Oregon State University "Mapping Parallel Iterative Algorithms onto Workstations Networks" A. Heddaya, K. Park, Boston University "A Task Graph Centroid" C. Leangsuksun, J. Potter Kent State University SESSION 6B: FAULT-TOLERANCE AND I/O IN DISTRIBUTED SYSTEMS Chair: I. Ahmad, Hong Kong University of Science & Technology "Data Reshuffling in Support of Fast I/O for Distributed-Memory Machines" C. Bornstein, P. Steenkiste Carnegie Mellon University "Process-Replication Technique for Fault-Tolerance and Performance Improvement in Distributed Computing Systems" J-F. Chiu, G-M. Chiu, National Taiwan Institute of Technology "Performance Evaluation of a Partial Dynamic Declustering Disk Array System" V. Catania, A. Puliafito, S. Riccobene, L. Vita, University of Catania, Italy 10:30 AM - 12:00 NOON Concurrent Sessions SESSION 7A: NETWORK PROTOCOLS AND INTERFACES Chair: I. Akyildiz, Georgia Institute of Technology "Large-Scale Group Communication Protocol on High-Speed Channel" M. Takamura, M. Takizawa, Tokyo Denki University "Deciding Boundedness for Systems of Communicating Finite State Machines" A. Benslimane, Universite de Franche-Comte, France "Design of the Header Processor for the PSi Implementation of the Logical Link Control Protocol in LANs" F. A. Morales, H. Abu-Amara, Texas A&M University SESSION 7B: PERFORMANCE EVALUATION AND RESOURCE MANAGEMENT Chair: J. J. Garcia-Luna, University of California, Santa Cruz "PMT: A Tool to Monitor Performance in Distributed Systems" V. Catania, O. Granato, A. Puliafito, L. Vita, University of Catania, Italy "Reducing Variations in Parallel Efficiency for Unstructured Grid Computations" B. Worner, G. Geunder, M. Hardtner, R. Zink, University of Stuttgart, Germany "Performance Prediction for Different Consistency Schemes in Distributed Shared Memory Systems" Z. G. Vranesic, S. Srbljic, University of Toronto L. Budin, University of Zagreb, Croatia ______________________________________________________________ KEYNOTE SPEAKERS Robert E. Kahn is President of the Corporation for National Research Initiatives (CNRI), which he founded in 1986 after a 13- year term at the Advanced Research Projects Agency (ARPA). Dr. Kahn holds a Ph.D. from Princeton University. Prior to his service at CNRI he held positions at Bell Laboratories, MIT and DARPA where he became Director of DARPA's Information Processing Techniques Office (IPTO) and initiated the United States government's billion dollar Strategic Computing Program, the largest computer research and development program ever undertaken by the federal government. Dr. Kahn is a member of the National Academy of Engineering, a Fellow of the IEEE, a Fellow of AAAI, a recipient of the AFIPS Harry Goode Memorial Award, the Marconi Award, the ACM SIGCOMM Award, the President's Award from the ACM, the IEEE Koji Kobayashi Computer and Communications Award, the ACM Software Systems Award, and the ASIS Special Award. He was twice the recipient of the Secretary of Defense Meritorious Civilian Service Award. Tilak Agerwala is Director of Architecture and System Design for IBM's parallel computer products. He is responsible for future product designs, performance analysis, and technical strategy. Dr. Agerwala has held executive positions in the RISC System 6000 division, where he was responsible for future system technology development, and at the T.J. Watson Research Center, where he initiated and directed broad research programs in parallel processing, artificial intelligence, and supercomputing. Dr. Agerwala is a member of the IBM Academy of Technology and has served on the Corporate Technical Committee. He was elected a Fellow of the IEEE for his leadership in the development of very high performance computers. ********************************************************************** DESCRIPTIONS OF TUTORIALS --------------------------------------------------------------------------- Tutorial 1A Software Systems and Tools for High Performance Distributed Computing Anand Tripathi, University of Minnesota The objective of this tutorial is to present an overview of the mechanisms and software tools/systems that are currently available for high-performance distributed programming in local-area networks. A high performance distributed programming environment needs mechanisms for specifying and managing parallel computation structures of a distributed application, transparent distribution of load in the network, dynamic linking of components, error detection and recovery, and transparency of the hardware/software heterogeneity in the network. This tutorial will present an overview of the most commonly used paradigms and models for distributed computing. These are related to interprocess communication models such as message passing and the remote procedure call (RPC), heterogeneous computing, load balancing and scheduling of tasks, and naming and protection of resources in the network. Reliability related issues in implementing RPC will be discussed. An overview the object model of computing will be presented in the context of micro-kernel architectures for distributed computing. The concept of group is a useful abstraction for managing a collection of related activities. Primitives for group management and broadcast communication will be presented. A number of software tools and programming languages are currently available for high performance distributed programming. An overview of the salient features and capabilities of Parallel Virtual Machine (PVM), Linda, Express, P4, CODE/ROPE, Mentat, CC++, PCN, Jade, Orca, Concurrent C, SR will be presented. SPEAKER: Anand Tripathi is an Associate Professor in the Department of Computer Science, University of Minnesota, Minneapolis. He received the B.Tech degree from the Indian Institute of Technology, Bombay, India, 1972, and the M.S. and Ph.D. degrees from the University of Texas at Austin in 1978 and 1980, respectively, in Electrical Engineering. From 1972 to 1975 he worked as a research scientist at Bhabha Atomic Research Center, Bombay, India. During 1981 to 1984 he worked as a Senior Principal Research Scientist at Honeywell Computer Sciences Center, Bloomington, MN. At the University of Minnesota he has led the design and development of the Nexus distributed operating system and its programming environment. His research interests include parallel and distributed systems, object-oriented programming, and fault-tolerant computing. In the past he has presented tutorials on distributed computing systems at some of the IEEE sponsored conferences and tutorial-weeks. --------------------------------------------------------------------------- Tutorial 1B Interfacing to High Speed Networks: Adapter Design and Operating System Issues K. K. Ramakrishnan, Digital Equipment Corporation Overview: We have seen significant increases in the bandwidth available for computer communication networks in the recent past. Commercially available Local Area Networks operate at 100 Mbits/sec and research networks are running at greater than 1 Gbit/sec. Host CPU processing speed has also increased relentlessly. However, the anticipation that end-user applications can effectively use these large communication link speeds has yet to be fulfilled to a large extent. Network I/O at the end-system has often been perceived as the bottleneck for distributed applications. This tutorial discusses the important topics in the design of a high-speed network adapter, including the memory-system architecture, buffer management, support for Quality of Service (QoS) guarantees and flow/congestion control, and host software issues. Several case studies will be presented. Speaker: K. K. Ramakrishnan is a Consulting Engineer in the Distributed Systems Architecture and Performance group at Digital Equipment Corporation. He has been with Digital since 1983 after completing his Ph.D. in Computer Science from the University of Maryland. His group is involved in architecting Digital's efforts in the networking and distributed systems area. Dr. Ramakrishnan has worked and published papers in the areas of load balancing, congestion control and avoidance, algorithms for FDDI, distributed systems performance and issues relating to network I/O. Dr. Ramakrishnan participates in the Internet Engineering Task Force and is a member of the End-End Group, as part of the Internet Research Task Force. He is also a technical editor for IEEE Network. --------------------------------------------------------------------------- Tutorial 2A Message Passing Using MPI: from Fundamentals to Applications David W. Walker, Oak Ridge National Laboratory This tutorial will describe the features of the MPI message passing standard, and will show how to use MPI in applications. The tutorial is intended to benefit researchers who have some experience with message passing, and who wish to assess the advantages of MPI for their particular applications. However, the tutorial will be structured to also be of use to novices. The tutorial will be divided into three main parts. The first part will give an overview of MPI, describe how it came about, and will discuss the basic point-to-point and collective communication capabilities of MPI. The second part will describe advanced features in MPI, in particular, the management of groups and communication contexts. The third part will be devoted to the presentation of application kernels and examples written using MPI. SPEAKER: David Walker is a Research Staff Member in the Mathematical Sciences Section at Oak Ridge National Laboratory, and an Adjunct Associate Professor in the Department of Computer Science of the University of Tennessee, Knoxville. He obtained his B.A. in Mathematics from Jesus College, Cambridge, in 1973, his M.S. in Astrophysics in 1979, and his Ph.D. in Physics from the University of London in 1983. From 1986-88. Dr. Walker was a member of the research staff of the Concurrent Computation Project at the California Institute of Technology, and from 1988-1990 was an Associate Professor in the Department of Mathematics at the University of South Carolina. He was worked at ORNL since 1990, where he is mainly involved in the design of software libraries, algorithms, and application for distributed memory concurrent computers. Dr. Walker was instrumental in establishing the MPI Forum, which led to the specification of the MPI standard and in which he played an active role. --------------------------------------------------------------------------- Tutorial 2B High Performance Distributed Computing in a Supercomputing Environment: Computational Services and Applications Issues W. T. C. Kramer and Horst D. Simon, NASA Ames Research Center The focus of this tutorial will be a discussion of current hardware and software trends for massively parallel supercomputers from the perspective of applications. In case studies the lessons learned at NASA Ames will be presented. The main thrust of the tutorial will be a presentation of high performance computing topics which will remain relevant for a long period of time, independent of currently ``hot" machines. This requires a detailed investigation of the issue of communication. From an applications perspective the communication characteristics of various applications will be examined. A new taxonomy of parallel application will be developed. Similarly high-performance architectures will be investigated with respect to their communications performance. The matching of the applications taxonomy with the architectural characteristics of a machine will form the basis for the understanding of high performance computing. High performance distributed computing will thus be evaluated as one possible alternative to both MPP and PVP systems. Emphasis will be placed on an honest evaluation of the capabilities and challenges that distributed computing faces in a production environment. The tutorial utilizes the unique experience made at the NAS facility at NASA Ames Research Center. Over the last five years at NAS massively parallel supercomputers such as the Connection Machines CM-2 and CM-5 from Thinking Machines Corporation and the iPSC/860 (Touchstone Gamma Machine) and Paragon Machines from Intel were used in a production supercomputer center alongside traditional vector supercomputers such as the Cray Y-MP and C90. In addition to these resources, NAS has operated since 1993 a distributed computing testbed (DCT), consisting of an environment with 50 SGI and 40 SUN workstations. Speakers: William Kramer is Chief of the NAS Computational Service Branch, and is responsible for providing state-of-the-art support and enhancement of a Cray YMP, two C-90s, CM-5, an Intel iPSC/860, an Intel Paragon and several Convex systems, along with over 250 other systems and workstations connected via Ultranet, FDDI and ethernet. These systems are used by a nationwide network of over 1400 researchers in 100+ sites, connected with the advanced NAS Aeronet and other networks. Bill was responsible for the first production Cray-2 and YMP at NAS before becoming Branch Chief. He is currently leading the NAS effort in computing on loosely clustered workstations. He has authored or presented papers sessions and seminars on a number of subjects, including system management, computer graphics, security and supporting advanced users at supercomputer centers. He and his staff presented the highest rated Tutorial at Supercomputing 89, "System Management in the UNIX Supercomputing Environment", which was presented again by invitation at SC '90. Horst D. Simon is with Computer Sciences Corporation at the Applied Research Branch at the NAS (National Aerodynamics Simulator) Systems Division at NASA Ames Research Center in Moffett Field, California. He is CSC department manager and responsible for a group of researchers in the areas of parallel algorithm development and scientific visualization, who as contractor personnel collaborate with the NAS staff. His research interests are in the development of high performance algorithms for vector and parallel machines. Particular areas of interest are sparse matrix algorithms, algorithms for large scale eigenvalue problems, and domain decomposition algorithms for irregular domains for parallel processing. Simon's algorithm research efforts were honored with the 1988 Gordon Bell Award for parallel processing research. Previously, until October 1989, Dr. Simon was an employee of Boeing Computer Services, and from October 1987 through October 1989, supported the NAS project in a similar capacity. From 1982 to 1983 Dr. Simon was an assistant professor at the Department of Applied Mathematics, SUNY Stony Brook, New York. Dr. Simon holds a Ph.D. in mathematics from the University of California (1982), Berkeley and a Diplom in Mathematik from the TU Berlin, Germany (1978). Dr. Simon has presented tutorials on high performance computing at all previous Supercomputing conferences. His tutorials have consistently been rated excellent or very good. --------------------------------------------------------------------------- CONFERENCE LOCATION HPDC-3 will be held at the Westin St. Francis Hotel in downtown San Francisco, California. Just 30 minutes from the airport, the Westin St. Francis is in the heart of the city on Union Square surrounded by fine restaurants, shops and theaters. Cable cars stop at the front door. The hotel is within walking distance of San Francisco's Chinatown and an easy cable car ride away from Ghiradelli Square and Fisherman's Wharf. Napa and Sonoma Valleys, the heart of California wine country, are an hour's drive away. If you have additional time before or after the symposium you may want to explore Marin County's redwoods, the California Mission, or drive scenic highway 101 to San Simeon. Monterey, Carmel, Santa Cruz, Yosemite National Park and Lake Tahoe are not far away. Weather: San Francisco in August is typically very pleasant. Sunny and warm (about 70F) during the day, and cool (about 50F) in the evenings. Note that the evening fog often makes it feel a bit colder and a lightweight jacket or all-weather coat is advised. Transportation: The hotel is about 20 minutes away from San Francisco International Airport. A taxi ride from the airport typically costs about $25. Also, the SFO Airporter has direct service to the hotel at 20-minute intervals. The fare is $10 one-way and $17 round-trip. Rental Cars: HERTZ has been appointed as the car rental supplier for HPDC-3. Special rates with unlimited mileage have been offered to HPDC-3 attendees and are also available the week prior and the week after the symposium. To make reservations call HERTZ at 1-800-654-2240 (In Canada 1-800-263-0600) and identify yourself as HPDC attendee (Meeting # 15329). Cars can be rented at most Bay-area airports and also at the Westin St. Francis Hotel. Social Event: An evening dinner-cruise on the San Francisco Bay is planned for Wednesday, August 3. The cost of the cruise is included in the registration fee. Additional tickets may be purchased at a price of $55 each. E-mail service will be available at the hotel to the attendees. ************************************************************************** REGISTRATION INFORMATION Advance registration is available using the form below through July 8th. E-mail registration is available only through July 27th, and must use a credit card number. On-site registration at the Westin St. Francis Hotel will be available starting Tuesday, August 2nd, and every day of the conference starting at 8 AM. --------------------------------------------------------------------------- HPDC-3 Registration Form Please send this form and a check, credit card information, or money order (no purchase orders) to the address below. Make checks payable to SYRACUSE UNIVERSITY. Registrations accepted via postal mail, FAX, or email only. SU Conference Planning HPDC-3 Conference P.O. Box 4709 Syracuse, NY 13221-4709 Phone: (315) 443-3333 Fax: (315) 443-1168 e-mail: hpdc@nova.npac.syr.edu ----------------------------------------------------------- Symposium Registration (select one): Advance Registration Regular Registration (Received by July 8, 1994) (Received after July 8, 1994) --------------------- ------------------- IEEE/ACM SIGCOMM Member [ ] $325 [ ] $395 Non-Member [ ] $405 [ ] $485 Full-time student [ ] $180 [ ] $180 Symposium registration fee includes a copy of the proceedings, sponsored lunches, coffee breaks, and the Bay cruise. Student registration does not include proceedings or the Bay cruise. Extra copies of the proceedings may be purchased on-site. ----------------------------------------------------------- Extra Dinner/Cruise Tickets ___ @ $55 each ----------------------------------------------------------- Tutorials (check the box for each tutorial attending) Note: You may select at most one tutorial from the morning session and one from the afternoon session. _____ 1A: Software Systems and Tools for High Performance | | Distributed Computing...............................|___| _____ 1B: Interfacing to High Speed Networks..................| | |___| _____ 2A: Message Passing Using MPI...........................| | |___| _____ 2B: High Performance Distributed Computing | | in a Supercomputing Environment.....................|___| Enter total number of tutorials at the appropriate rate (rates are per tutorial). Advance Registration Regular Registration (Received by July 8, 1994) (Received after July 8, 1994) --------------------- ------------------- IEEE/ACM SIGCOMM Member ___ @ $165 ___ @ $195 Non-Member ___ @ $205 ___ @ $245 Total Tutorial Fee: $________US Each Tutorial registration fee includes attendance at one tutorial session, notes, and coffee breaks. There are no student fees for the tutorials. Cancellations of tutorial registrations after July 22 will be subject to the total fee. We reserve the right to cancel the tutorials due to insufficient participation or other unforeseeable problems, in which case fees will be refunded. ----------------------------------------------------------- Symposium Registration Fee $_________ + Extra cruise tickets $_______ + Tutorial Fees $___________ = Total Amount enclosed: $________ US ----------------------------------------------------------- Name: ____________________________________ Affiliation: _____________________________ Address: _________________________________ City/State/Zip/Postal Code/Country: ________________________________ Phone: ___________________________________ FAX: _____________________________________ E-mail Address: ___________________________ IEEE/ACM SIGCOMM Member #: ____________________________ (required for member rate) Method of Payment __ Visa __ Mastercard __check enclosed (payable to Syracuse University) Card Holder Name: ________________________ (as it appears on card) Card Number: _____________________________ Expiration Date: _________________________ Signature: _______________________________ You may register by e-mail using your credit card. Send completed registration to hpdc@nova.npac.syr.edu If you have a disability and may require accommodation in order ______ to fully participate in this activity, please check | | here. You will be contacted to discuss your needs. |____| Vegetarian Meals: ___ YES / ___ NO Cancellation Policy: All requests for refunds must be received in writing on or before July 22, 1994. No refunds will be made on cancellations made after this date. ************************************************************************** HOTEL REGISTRATION Please contact the Westin St. Francis Hotel directly for accommodations. The Westin St. Francis Union Square 335 Powell Street San Francisco, CA 94102 USA Phone: (415) 397-7000 Special conference rates have been arranged for attendees of HPDC-3. The rates are as follows: Type of Room Main Building Towers Single Double Single Double Standard $120 $120 - - Medium $135 $135 - - Deluxe $150 $150 $170 $170 Reservation cut-off date is Friday, July 1, 1994; Reservations made after this date will be on a space-available basis only. In order to receive our special rates you must identify yourself as a participant in the HPDC-3 Conference. ************************************************************************** CONFERENCE COMMITTEES ************************************************************************** SYMPOSIUM GENERAL CHAIR: Geoffrey Fox, NPAC, Syracuse University SYMPOSIUM STEERING COMMITTEE: - Salim Hariri, Syracuse University (Chair) - Tilak Agerwala, IBM - Andrew Grimshaw, University of Virginia - H. T. Kung, Harvard University - T. V. Lakshman, Bell Communications Research - Daniel McAuliffe, Rome Laboratory - C. S. Raghavendra, Washington State University PROGRAM CHAIR: Anujan Varma, University of California, Santa Cruz PUBLICITY CHAIRS: North America: T. V. Lakshman, Bell Communications Research Europe: Walid Dabbous, INRIA, France Asia: Makoto Takizawa, Tokyo Denki University, Japan TUTORIALS CHAIR: Ian F. Akyildiz, Georgia Tech EXHIBITS: C. S. Raghavendra, Washington State University REGISTRATION AND LOCAL ARRANGEMENTS: Peggy Van Arnam, Syracuse University PROGRAM COMMITTEE - Dharma Agrawal, North Carolina State University - Prathima Agrawal, AT&T Bell Labs - Ian F. Akyildiz, Georgia Tech - Marco Annaratone, DEC - Ken Birman, Cornell University - Suresh Chalasani, University of Wisconsin, Madison - Monsong Chen, IBM Research - Roger Chen, Syracuse University - Abdelaziz Chihoub, Siemens Corporate Research - Jon Crowcroft, University College London - Walid Dabbous, INRIA, France - Patrick Dowd, SUNY Buffalo - Dennis Duke, SCRI/Florida State University - Stuart Elby, NYNEX Science and Technology - Richard Freund, NRaD - J.J. Garcia-Luna, University of California, Santa Cruz - Arif Ghafoor, Purdue University - Andrew Grimshaw, University of Virginia - Salim Hariri, Syracuse University - S. H. Hosseini, University of Wisconsin, Milwaukee - T. V. Lakshman, Bell Communications Research - C. R. Mechoso, UCLA - Paul Messina, Caltech - Richard C. Metzger, Rome Laboratory - Paul Mockapetris, USC/ISI - John Morrison, Los Alamos National Laboratory - John Nicholas, Battelle Pacific Northwest Lab - James C. Patterson, Boeing Co. - Ira Pramanick, IBM - Michael Quinn, Oregon State University - C. S. Raghavendra, Washington State University - Paul Rupert, Lawrence Livermore National Laboratory - Karsten Schwan, Georgia Institute of Technology - Nita Sharma, Ncube Inc. - Vaidy Sunderam, Emory University - Makoto Takizawa, Tokyo Denki University, Japan - Alexander Thomasian, IBM Research - Satish Tripathi, University of Maryland - Larry Wittie, SUNY Stony Brook - Pen-Chung Yew, University of Illinois