CANPC '98 Advance Program (1/31, Las Vegas

                                                            Advance Program
                                    (http://www.cis.ohio-state.edu/~canpc98/)


  Workshop on Communication, Architecture, and Applications for Network-based
                                                          Parallel Computing


                                                                    CANPC '98


        to be held in conjunction with Fourth International Symposium on High
                                  Performance Computer Architecture (HPCA-4)
                                              (http://www.cs.hku.hk/~hpca4/)
            Orlean's Hotel, Las Vegas, Nevada, January 31 - February 1, 1998


                                      Sponsored by the IEEE Computer Society


----------------------------------------------------------------------------
    Deadline for Advance Registration and Hotel Reservation is January 15th.
----------------------------------------------------------------------------


                                                  Preliminary Advance Program


January 31 (Saturday)


  9:00 - 12:00 Registration


12:00 - 12:10 Opening


12:10 - 1:15 Keynote Speech


          Bob Horst, Tandem Computers
          Key Issues in Designing System Area Networks


  1:15 - 1:30 Break


          Each session involves presentation by the authors as well as group
                                                      discussion at the end.


  1:30 - 3:30 Session I


          Network Interface and Communication Protocol
          Session Chair: Kourosh Gharachorloo, DEC WRL


                * The Remote Enqueue Operation on Networks of Workstations, E.
                    P. Markatos, M. G. H. Katevenis, and P. Vatsolaki, Institute
                    of Computer Science, FORTH, Greece
                * The HAL Interconnect PCI Card, J. Larson, HAL Computer
                    Systems
                * Implementing Protected Multi-User Communication for Myrinet,
                    S. Nagar, D. Seed, and A. Sivasubramaniam, The Pennsylvania
                    State University
                * ECOLE: A Configurable Environment for a Local Optical Network
                    of Workstations, G. Brebner and R. Pooley, Univ. of
                    Edinburgh, Scotland


  3:30 - 4:00 Coffee Break


  4:00 - 6:00 Session II


          Systems, Application and Application Support
          Session Chair: Jaswinder Pal Singh, Princeton University


                * The Design of a Parallel Programming System for a Network of
                    Workstations: An Object-Oriented Approach, C. W. Ming, S.
                    Chanson, and M. Hamdi, The Hong Kong University of Science
                    and Technology
                * Remote Subpaging Across a Fast Network, by M. Bangalore and
                    A. Sivasubramaniam, The Pennsylvania State University
                * Improved Functional Imaging through Network Based Parallel
                    Processing, F. Munz, T. Stephan, U. Maier, T. Ludwig, A.
                    Bode, S. Ziegler, S. Nekolla, P. Bartenstein, and M.
                    Schwaiger, Technische Universitat Munchen (TUM), Germany
                * AutoMap and AutoLink: Tools for Communicating Complex and
                    Dynamic Data-structures Using MPI, D. S. Goujon (Telecom
                    INT), M. Michel (RESEDAS), J. Peeters (Univ. of Twente), and
                    J. E. Devaney (NIST)


February 1 (Sunday)


  8:00 - 8:30 Registration


  8:30 - 10:00 Session III


          Communication and Synchronization
          Session Chair: Jose Duato, Universidad Politecnica de Valencia


                * Analysis of a Programmed Backoff Method for Parallel
                    Processing on Ethernets, N. Matloff, University of California
                    at Davis
                * Improving Dynamic Token-Based Distributed Synchronization
                    Performance via Optimistic Broadcasting, A. I-C. Lai and
                    C.-L. Lei, National Taiwan University, Taiwan
                * Fast Barrier Synchronization on Shared Fast Ethernet, G.
                    Chiola and G. Ciaccio, Universita di Genova, Italy


10:00 - 10:30 Coffee Break


10:30 - 12:00 Session IV


          Routing and Flow-control
          Session Chair: Manolis Katevenis, FORTH and Univ. of Crete, Greece


                * Parallel Routing Table Computation for Scalable IP Routers,
                    X. Xiao and L. M. Ni, Michigan State University
                * A Tool for the Analysis of Reconfiguration and Routing
                    Algorithms in Irregular Networks, R. Casado, M. B. Caminero,
                    P. Cuenca, F. J. Quiles, A. Garrido (Universidad de Castilla
                    - La Mancha), and J. Duato (Universidad Politecnica de
                    Valencia)
                * Real-Time Traffic Management on Optical Networks using
                    Dynamic Control of Cycle Duration, C.-C. Yu and S.
                    Bhattacharya, Arizona State University


12:00 - 2:00 Lunch Break


  2:00 - 4:00 Session V


          Communication Patterns and Performance Evaluation
          Session Chair: Craig Stunkel, IBM T. J. Watson Research Center


                * A Comparative Characterization of Communication Patterns in
                    Applications using MPI and Shared Memory on an IBM SP2, S.
                    Karlsson and M. Brorsson, Lund University, Sweden
                * Characterization of Communication Patterns in Message-Passing
                    Parallel Scientific Application Programs, J. Kim and D. J.
                    Lilja, University of Minnesota
                * Cross-platform Analysis of Fast Messages for Myrinet, G.
                    Iannello, M. Lauria, and S. Mercolino, Universita di Napoli
                    Federico II, Italy
                * Performance of a Cluster of PCI Based UltraSparc Workstations
                    Interconnected with SCI, K. Omang (University of Oslo)


  4:00 - 4:30 Coffee Break


  4:30 - 6:00 Panel Session


          Strategies for NOW Performance Scaling: Shared-nothing or
          Shared-something?


          Moderator: Greg Pfister, IBM


          Panelists: TBA


----------------------------------------------------------------------------


                                          Registration and Hotel information


Workshop registration is handled by the HPCA-4 conference registration
form (http://www.ee.iastate.edu:80/~prasant/hpca/). The workshop
registration fees are as follows. The registration fee includes a copy
of the CANPC '98 proceedings, published by Springer-Verlag. Sign up
by January 15th for lowest rates. You can sign up for just the CANPC
'98 workshop, or a combination of HPCA-4 and CANPC'98.




                                                                                              CANPC'98 Workshop


                                                                                    Advance On-site


  IEEE Members $90 $110


  Non-members $110 $135


  Full-time Students $60 $70


For hotel reservation information, see the HPCA-4 information page
(http://www.cs.hku.hk/~hpca4/). Sign up by January 15th to receive the
lowest rates and to guarantee room availability. The conference rate for
Friday and Saturday (both nights together) arrival is $89/night and for
Sun-Tue (if you are attending HPCA-4) is $39/night . Reservation can be made
by directly calling the Orlean's Hotel at (800) 675-3267.


----------------------------------------------------------------------------


THEME:


Clusters of workstations connected by local area networks (LANs) have gained
popularity as a platform for cost-effective parallel processing,
establishing the paradigm of ``network-based parallel computing.'' Hardware
and software LAN technology was not initially developed for parallel
processing, and thus the communication overhead between workstations can be
quite high. This has forced severe constraints on maximizing parallel
performance in workstation clusters.


A large number of research groups from academia, industry, and research labs
are currently engaged in deriving novel hardware and software solutions to
alleviate these bottlenecks. Many new interconnection technologies (such as
Fast Ethernet, ATM, Myrinet, Fibre Channel), software schemes (such as Fast
Message, U-Net, Active Message), and standards (such as PacketWay) are being
developed to provide low-latency and high-bandwidth interconnection for
network-based parallel computing. These developments are facilitating the
migration of many parallel applications to network-based platforms
consisting of PCs/Workstations.


The goal of this workshop is to bring together researchers and practitioners
working in the areas of communication, architecture, and applications to
discuss state-of-the-art solutions as well as future trends for designing
cost-effective network-based parallel computing systems.


TOPICS OF INTEREST:


Topics of interest include, but not limited to: architectural support for
achieving fast communication and synchronization; network interface design;
implementation of efficient messaging layers; LAN interconnection with
high-speed switches and routing; characterization of communication and
synchronization traffic; collective communication support in both software
and hardware; experience in implementing MPI and HPF; supporting shared
memory paradigms; characterizing communication and synchronization
requirements in applications; emerging network-based parallel computing
applications; interplay between hardware/software support and application
characteristics; and performance evaluation and benchmarking. Results of
both theoretical and practical significance will be considered.


PROCEEDINGS:


The proceedings will be published as one issue of Springer-Verlag's
``Lecture Notes in Computer Science'' (LNCS) series.


WORKSHOP CO-CHAIRS:


Dhabaleswar K. Panda (Ohio State) and Craig B. Stunkel (IBM T.J. Watson)


PROGRAM COMMITTEE:


      * Henri Bal (Vrije U., The Netherlands)
      * Prith Banerjee (Northwestern Univ.)
      * Debashis Basak (Fore Systems)
      * Jehoshua Bruck (Caltech)
      * Andrew Chien (Univ. of Illinois)
      * Al Davis (Utah)
      * Jose Duato (U. Politecnica de Valencia)
      * Sandhya Dwarkadas (Rochester)
      * Thorsten von Eicken (Cornell)
      * Michael Foster (NSF)
      * Mike Galles (SGI)
      * Kourosh Gharachorloo (DEC WRL)
      * Bob Horst (Tandem Computers)
      * Vipin Kumar (Minnesota)
      * Andreas Nowatzyk (DEC WRL)
      * Greg Pfister (IBM)
      * Vernon Rego (Purdue)
      * Jaswinder Pal Singh (Princeton)
      * Marc Snir (IBM T.J. Watson)
      * Per Stenstrom (Chalmers U., Sweden)
      * Anand Tripathi (Minnesota)
      * Wolf-Dietrich Weber (HAL Computers)
      * Sudhakar Yalamanchili (Georgia Tech)


----------------------------------------------------------------------------


ADDITIONAL INFORMATION:


For further questions, send e-mail to canpc98@cis.ohio-state.edu.
--

Return to the comp.compilers page.
Search the comp.compilers archives again.