From: craig@puma.cat.syr.edu (Gary L. Craig) Subject: CONF: High Performance Dist. Computing (early reg. deadline - 8/18) Date: 11 Aug 92 00:01:46 GMT Organization: Syracuse University Crossposted from news.announce.conferences First International Symposium on High Performance Distributed Computing Wednesday September 9 - Friday September 11, 1992 Syracuse, New York Location: Sheraton University Inn & Conference Center, 801 University Avenue, P.O. Box 801, Syracuse, NY 13210-0801 Tel: (315) 475-3000 SPONSORS: IEEE Computer Society The New York State Center for Advanced Technology in Computer Applications and Software Engineering (CASE) at Syracuse University Northeast Parallel Architectures Center (NPAC) at Syracuse University IN COOPERATION WITH: ACM SIGCOMM The theme of this symposium is to investigate software techniques and architectural support for application of parallel and distributed computing for solving computationally intensive applications across a network of high-performance computers. The papers to be presented address all aspects of achieving parallel and distributed computing across a high speed network of computers. Exhibits, prototypes, software tools and applications will also be shown and presented during the symposium. SYMPOSIUM CHAIR: Geoffrey Fox, NPAC, Syracuse University, (315) 443-4741, hpdc@nova.npac.syr.edu PROGRAM COMMITTEE CHAIR: Salim Hariri, Syracuse University, (315)443-4282, hariri@cat.syr.edu PROGRAM COMMITTEE - Dharma Agrawal, North Carolina State University - Ken-ichi Aihara, NTT Telecommunication Networks Labs, Japan - Alok Choudhary, Syracuse University - Dennis Duke, SCRI/Florida State University - Arif Ghafoor, Purdue University - Jose Joaquin Garcia-Luna, SRI International - Andrew Grimshaw, University of Virginia - Kai Hwang, University of Southern California - Carl Kesselman, Caltech - Paul Mockapetris, DARPA - C. S. Raghavendra, Washington State University - Anthony Reeves, Cornell University - Vaidy Sunderam, Emory University - Muharrem Uyar, AT&T - Jon Valente, Rome Labs - Anujan Varma, Univ. California, Santa Cruz - Benjamin W. Wah, University of Illinois - Gregory F. Wetzel, AT&T - Gordon Wright, IBM PUBLICITY: Anujan Varma, University of California, Santa Cruz, (408) 459-3505, varma@cse.ucsc.edu FINANCE and EXHIBITS CHAIR: A. Gaber Mohamed, NPAC, Syracuse University, (315)443-1722, agm@nova.npac.syr.edu LOCAL ARRANGEMENTS: Gary Craig, Syracuse University, (315) 443-4389, craig@cat.syr.edu ---------------------------------------------------------- Preliminary Program Wednesday, September 9, 1992 7:00-10:00 am Registration 9:00-9:15 am Welcome Address 9:15-10:15 am Keynote Speech: H. T. Kung, High-Performance Computing over a Network Backplane 10:15-10:30 am Break 10:30-Noon SESSION 1: Invited Papers Chair: Jose Joaquin Garcia-Luna, SRI International "Issues in Networking and Data Management of Distributed Multimedia Systems" by P. B. Berra, C-Y.R. Chen, Syracuse University, A. Ghafoor, Purdue University, and T.D.C. Little, Boston University "Distributing Climate Models over Gigabit Networks" by C. Roberto Mechoso, University of California at Los Angeles Noon-1:30 pm Lunch 1:30-3:30 SESSION 2: Tools and Programming Paradigm Chair: Carl Kesselman, Caltech "Data-Parallel Programming on a Network of Heterogeneous Workstations" by M. Quinn, N. Nedelijkovic, Oregon State University "MOVIE Model for Open Systems based High Performance Distributed Computing" by C. Faigle, W. Furmanski, T. Haupt, J. Niemiec, M. Podgorny, D. Simoni, NPAC, SU "No Pain and Gain - Experiences with Mentat on a Biological Application" by A. Grimshaw, E. West, W. Pearson, University of Virginia, Charlottesville "Experiences in High Performance Computing with Pleiades and ESP" by L. Wilson, M. Gonzalez, M. Cruess, University of Texas at Austin 1:30-3:30 SESSION 3: High Speed Networks and Protocols Chair: Dennis Duke, SCRI/Florida State University "Parallel Networking and Visualization on the Connection Machine CM-5" by Gary Oberbrunner, Thinking Machines Corp "A Scalable Packet Switch for Distributed Computing" by A. Guha, M. Agrawal, Honeywell "Performance Comparison of Reservation and Preallocation Protocols for Star-Coupled WDM Photonic Networks" by P. Dowd, K. Bogineni, SUNY Buffalo "Determining Update Latency Bounds in Galactica Net" by S. Clayton, R. Duckworth, W. Michalson, A. Wilson, Worcester Polytechnic Institute 3:30-4:00 pm Break 4:00-5:30 pm SESSION 4: Applications I Chair: Richard Freund, Naval Ocean Systems Center "Harmonizing a Distributed Operating System with Parallel and Distributed Applications" by Y. Sinjo, Y. Kiyoki, University of Tsukuba, Japan "A Radiative Heat Transfer Simulation on a SPARCStation Farm" by R. Minnich, D. Pryor, Supercomputing Research Center "Particle Simulation on Heterogeneous Distributed Supercomputers" by J. Becker, University of California, Davis and L. Dagum, NASA Ames Research Center 4:00-5:30 pm SESSION 5: Requirement Analysis and Architectural Support for HPDC Chair: Gurudatta Parulkar, Washington University "Requirement Analysis for High Performance Distributed Computing over LAN's" by Manish Parashar, S. Hariri, G. Mohammad, G. Fox, NPAC, Syracuse University "High Performance Computing on a Cluster of Workstations" by A. Cheung , A. Reeves, Cornell University "High Performance PDU Processing for Application Layer" by B. Sarikaya, Bilkent University, Turkey and M. Bilgic, Concordia University, Canada 6:00-8:00 pm BUFFET, Demonstrations, and Bird-of-a-Feather - The activities will be held at Syracuse University campus, a walking distance from the Sheraton Hotel. Thursday, September 10, 1992 9:00-10:00 Keynote Speech, TBA 10:00-10:30 am Break 10:30-noon SESSION 6: Multimedia Chair: Kamal Jabbour, Syracuse University "A Flexible and Adaptive Transport System Architecture to Support Multimedia Applications on High-Speed Networks" by D. Schmidt, D. Box, T. Suda, University of California, Irvine "FDDI Support for Multimedia Traffic" by F. Schaffa, B. Patel, M. Willebeek-LeMair, IBM T.J. Watson Research Center "On Flow Control in Multimedia Networks" by P. Varshney and S. Dey, Syracuse University 10:30-noon SESSION 7: Load Balancing Chair: C.S. Raghavendra, Washington State University "Physical-Level Synthetic Workload Generation for Load Balancing Experiments" by P. Mehra, B. Wah, University of Illinois "Performance Comparison of Active-Sender and Active-Receiver Policies for Distributed Caching" by C. Pu, D. Florissi, P. Soares, K.L. Wu, P.S. Yu, Columbia University, IBM T.J. Watson Research Center "Load Distribution on Hypercube Systems Supporting Applications with Dynamic Structures" by I. Ahmad, G. Fox, Syracuse University, A. Ghafoor, Purdue University Noon-1:30 pm Lunch 1:30-3:00 pm SESSION 8: Panel Session: "Requirements and Directions for High Performance Distributed Computing" Chair: Malvin H. Kalos, Cornell University Panelists: 1. Dick Metzger, Rome Labs 2. B. Gopinath, Rutgers University 3. Tilak Agerwalla, IBM Austin 4. Justin Rattner, Intel Corp, 5. Marco Annaratone, Digital Corp. 3:00-3:20 pm Break 3:20-4:20 pm SESSION 9: Applications II Chair: Jim Gordon, IBM - Cornell Theory Center "Lattice Boltzmann Method on a Cluster of IBM RISC System/6000" by G. Betello, G. Richelli, S. Succi, IBM ECSEC, F. Ruello, University of Catania "Computing Radiosity on a High Performance Workstation LAN" by G. Singh, F. Westervelt, Wayne State University, S. Abraham, University of Michigan 3:20-4:20 pm SESSION 10: Distributed System Issues Chair: John Peterson, JPL "Maintaining Good Performance In Disk Arrays During Failure via Uniform Parity Group Distribution" by S. Ng, R. Mattson, IBM Almaden "A Distributed Scheduling Simulation" by D. Baker, K. Wika, R. Haddleton, University of Virginia Friday, September 11, 1992: Post Symposium Tutorials Time: 9:00am 5:00pm Tutorial I: (full day) Paradigms and Tools for Heterogeneous Network Computing Vaidy Sunderam, Emory University This tutorial will cover issues, methodologies, and strategies for concurrent computing on heterogeneous networks of independent computer systems. We will begin with an overview of software systems and tools that are available to support network-based computing. A descriptive outline of several production applications in a variety of disciplines will be presented to demonstrate the effectiveness and viability of network computing. The remainder of the course will be devoted to understanding PVM and HeNCE -- software systems that enable concurrent computing on heterogeneous collections of multiprocessors, supercomputers, scalar machines, and workstations. PVM (Parallel Virtual Machine) is the software infrastructure that allows heterogeneous groups of machines to be used as a general-purpose concurrent computing resource. The tutorial will focus on developing concurrent applications for PVM, using several models of parallelism. HeNCE is a graphical toolkit and methodology that significantly eases the task of application development for PVM. Developers specify applications using such a graph, and associate predefined program modules with different points in the graph. The HeNCE toolkit then executes the graph on a user-configured virtual machine (under PVM). BIO: Vaidy Sunderam received a PhD in Computer Science from the University of Kent, England, and is a faculty member in the department of Math & Computer Science at Emory University, Atlanta, USA. His research interests are in parallel and distributed processing, particularly high-performance concurrent computing in heterogeneous networked environments. Sunderam's recent research has focused on heterogeneous concurrent computing on general purpose networks. He is the principal architect of the PVM system for concurrent heterogeneous computing, which is in widespread use and is emerging as a de-facto standard for network-based heterogeneous systems. He is the recipient of the 1990 IBM supercomputing first prize award for his work on high-performance, network concurrent computing. His other recent research includes high-speed protocols for distributed systems support, graphical tools for parallel program development, concurrent stochastic simulation, and algorithms for efficient implementation of concurrency primitives. Tutorial II: (half day) Time: 2:00pm - 5:00pm Multimedia Information Systems P. Bruce Berra - CASE Center, Syracuse University Multimedia information systems are receiving increased attention in business, industry, and government. These systems integrate the use of such diverse data types as voice, text, numeric, image, graphic and video. While this field is still in its infancy, limited commercial products are beginning to appear and there is considerable interest in the research issues that need to be addressed before fully integrated systems can become a reality. In this tutorial we will discuss many of the multimedia issues in the following contexts: a)Multimedia background - types, issues, etc.; b) Applications - medical, office automation, entertainment, manufacturing, user interface systems; c) Data modeling and management - query languages, multimedia data integration and synchronization, object oriented approaches; d) Data storage and access - storage of images and video, data compression, multimedia access; e) Networks and distributed systems - network requirements, concurrency control, multidatabase issues; f) Future research and development issues and areas. BIO: P. Bruce Berra is currently professor of Electrical and Computer Engineering and a member of the faculty of the Computers and Information Science at Syracuse University. He is director of the New York State Center for Advanced Technology in Computer Applications and Software Engineering (CASE). He is also president of PBB systems, an AI/database consulting firm. His research interests include multimedia information systems, parallel processing for very large data and knowledge bases and optical database machines. Prof. Berra received the BS and MS degress from the University of Michigan and the PH.D. from Purdue University. He is a Fellow of the IEEE. Tutorial III: (half day) Time: 9:00am - 12:00pm Distributed Shared Memory Systems Prof. Larry Wittie - Computer Science, SUNY Stony Brook Distributed Shared Memory (DSM) mechanisms provide the parallel programming convenience of globally shared memory, without the hardware costs and scaling limits of traditional shared memory multiprocessors. They let computers in a network share selected portions of data to solve a single application problem rapidly, in parallel. Invisible DSM messages share needed data consistently among processor caches or memories. DSM systems have been implemented both in software (IVY, Munin) for user convenience and development of ideas, and in hardware (DASH, SCI, Sesame) for high-performance parallel computing on networks. DSM mechanisms range in a spectrum from on-demand fetch to eager sharing. Most coherent cache multicomputers (Multicube, DASH) and software DSM systems use demand fetch, which does not run most single programs efficiently ("scale") on more than a few hundred processors simultaneously. Eager sharing hardware (Sesame) hides most network communication delays during data sharing and provides efficient parallel execution upon thousands of processors. This tutorial will cover DSM data consistency, latency hiding, and other critical implementation issues; summarize representative systems; and compare effective execution rates on huge networks. Latency-hiding DSM mechanisms will be incorporated within cache coherence protocols for future massively parallel, high-performance computing systems. DSM will simplify teraflops computing for many applications. BIO: Larry Wittie is a professor of Computer Science at the State University of New York at Stony Brook and previously taught at Purdue and at SUNY/Buffalo. He received his Ph.D. degree in Computer Science from the University of Wisconsin at Madison in 1973. He is an editor for IEEE Transactions on Parallel and Distributed Systems and a member of IEEE, ACM, the Society for Neuroscience, and Sigma Xi. He currently is working on reliable eagerly shared distributed memory interfaces, including gigabit network hardware, reliable ordered multicast protocols, and software tools for parallel programming. He is interested in the organization of large distributed systems, including interconnections, operating systems, and debugging for large multicomputers. His long term interests lie in efficient ways to build and control networks of millions of computers. Prof. Wittie's addresses are lw@sbcs.sunysb.edu and Department of Computer Science, State University of New York, Stony Brook, NY USA 11794-4400. CONFERENCE/TUTORIAL REGISTRATION FORM: ______________________________________________________________________ Early registration for either the conference or the tutorial must be received by August 15, 1992. Payment must be remitted by check or money order in U.S. currency only, payable to Syracuse University ------------------- Please circle applicable charges. ________________________________________________________________ | |Conference Registration| Tutorial Registration | | |_______________________|_____________________________| | |Advance |Late/On Site |Advance |Late/On Site | | |by Aug 18|after Aug 18 |by Aug. 18 |after Aug 18 | | | | | | | | | | | 1 2/3 both| 1 2/3 both| |_________|_________|_____________|__________2+3_|__________2+3_| |IEEE/ACM | | | | | | Member |$295 |$355 |$200 $150 $225|$230 $180 $255| |Nonmember|$370 |$445 |$235 $185 $260|$265 $215 $290| |Student* |$150 |$150 |$200 $150 $225|$230 $180 $255| |_________|_________|_____________|______________|______________| Enter Tutorial(s) (1, 2, 3 or 2+3) you wish to register for:______ Total Enclosed $__________ Please return registration form and fee to: Donna L. McCammon CASE Center 2-212 Science & Technology Center Syracuse University Syracuse, NY 13244-1190 Phone: (315)443-1066 Email: donna@cat.syr.edu Fax: (315)443-4745 PLEASE TYPE OR PRINT Name _________________________________ Company __________________________________ Address/Mailstop __________________________________________________ City/State/Zip/Country ____________________________________________ Email address, if available _______________________________________ IEEE/ACM Membership Number (required for member rate) ______________ _____________________________________________________________________ ROOM RESERVATION INFORMATION : The Sheraton University Inn & Conference Center located adjacent to the main campus of Syracuse University and close to downtown Syracuse. Directions: The Conference Center is located 8 miles from Syracuse Airport, and can be reached by taxi, rental car and shuttle service. >From the airport, take I-81 South to the Harrison Street exit (stay left following signs to Syracuse University -- on Almond Street). Turn left at the 2nd light onto Adams Street. At the 4th stop light turn right on University Avenue. The Sheraton Inn is 1.5 blocks on your left. ______________________________________________________________________ |HOTEL RESERVATIONS | |Sheraton University Inn - (315) 475-3000 | |Rates: Single - $89.00; Double - $96.00 | |The cutoff date for reservations at these special rates is TUESDAY | |AUGUST 18. Please call your reservation requests and | |mention IEEE COMPUTER SOCIETY SPECIAL RATES FOR HPDC. | | | |Genesee Inn Executive Quarters - (800) 365-HOME | |1060 E. Genesee St. (corner of University Ave. 4.5 blocks from conf.)| |Rates: Single - $52.00; Double - $69.00 | |_____________________________________________________________________| -- Gary L. Craig email: craig@cat.syr.edu Dept. ECE - Syracuse University phone: (315) 443-4389 121 Link Hall Syracuse, NY 13244-1240 FAX: (315) 443-2583 (or x4745)