TUESDAY, APRIL 13

9:00 AM - 10:00 AM
KEYNOTE ADDRESS

The Changing Nature of Concurrent Computing Research
K. Mani Chandy, California Institute of Technology


10:30 AM - 12:30 PM
SESSION 1
Algorithmic Paradigms and Primitives
Chair: Ernst Mayr
Technische Universitaet Muenchen, Germany


The Characterization of Data-Accumulating Algorithms
Stefan D. Bruda and Selim G. Akl, Queen's University, Canada

Prefix Computations on Symmetric Multiprocessors
David R. Helman and Joseph Ja'Ja', University of Maryland

Reducing I/O Complexity by Simulating Coarse Grained Parallel Algorithms
Frank Dehne, Carleton University, Canada, Wolfgang Dittrich, Bosch Telecom, Germany, David Hutchinson and Anil Maheshwari, Carleton University, Canada

Lower Bounds on the Loading of Degree-2 Multiple Bus Networks for Binary-Tree Algorithms
Hettihe P. Dharmasena and Ramachandran Vaidyanathan, Louisiana State University

A Time-Optimal Solution for the Path Cover Problem on Cographs
K. Nakano, Nagoya Institute of Technology, Japan, S. Olariu, Old Dominion University, A.Y. Zomaya, University of Western Australia, Australia

Parallel Matrix Multiplication on a Linear Array with a Reconfigurable Pipelined Bus System
Keqin Li, State University of New York at New Paltz, Victor Y. Pan, Lehman College, CUNY

10:30 AM - 12:30 PM
SESSION 2
Latency Tolerance and Performance Modeling
Chair: Jose' Duato
Universidad Politecnica de Valencia, Spain


Improving Collective I/O Performance Using Threads
Phillip M. Dickens, Illinois Institute of Technology, Rajeev Thakur, Argonne National Laboratory

Linear Aggressive Prefetching: A Way to Increase the Performance of Cooperative Caches
T. Cortes and J. Labarta, Universitat Politecnica de Catalunya, Spain

Hiding Communication Latency in Reconfigurable Message-Passing Environments
Ahmad Afsahi and Nikitas J. Dimopoulos, University of Victoria, Canada

Performance Impact of Memory Hierarchies and Networks on Cluster Computing
Xing Du, University of Virginia, Xiaodong Zhang, College of William and Mary

A Factorial Performance Evaluation for Hierarchical Memory Systems
Xian-He Sun and Dongmei He, Louisiana State University, Kirk W. Cameron, Louisiana State University and Los Alamos National Laboratory, Yong Luo, Los Alamos National Laboratory

A Performance Model of Speculative Prefetching in Distributed Information Systems
N.J. Tuah, M. Kumar, and S. Venkatesh, Curtin University of Technology, Australia

10:30 AM - 12:30 PM
SESSION 3
Communication, Run-Time Systems
Chair: Calvin Lin
University of Texas at Austin



Run-Time Selection of Block Size in Pipelined Parallel Programs
David K. Lowenthal and Michael James, University of Georgia

Reducing Parallel Overheads Through Dynamic Serialization
Michael J. Voss and Rudolf Eigenmann, Purdue University

Using Channels for Multimedia Communication
David May and Henk L. Muller, University of Bristol, United Kingdom

The Paderborn University BSP (PUB) Library -- Design, Implementation and Performance
Olaf Bonorden, University of Paderborn, Germany, Ben Juurlink, Delft Technical University, The Netherlands, Ingo von Otto and Ingo Rieping, University of Paderborn, Germany

A Capabilities Based Communication Model for High-Performance Distributed Applications: The Open HPC++ Approach
Shridhar Diwan and Dennis Gannon, Indiana University


2:00 PM - 4:00 PM
SESSION 4
Scalable Computing
Chair: Gianfranco Bilardi
Università di Padova and University of Illinois


Average-Case Analysis of Isospeed Scalability of Parallel Computations on Multiprocessors
Keqin Li, State University of New York at New Paltz, Xian-He Sun, Louisiana State University

Fully-Scalable Fault-Tolerant Simulations for BSP and CGM
Sung-Ryul Kim and Kunsoo Park, Seoul National University, Korea

Coarse Grained Parallel Maximum Matching in Convex Bipartite Graphs
J. Bose, A. Chan, F. Dehne, and M. Latzel, Carleton University, Canada

Experimental Evaluation of QSM, a Simple Shared-Memory Model
Brian Grayson, Michael Dahlin, and Vijaya Ramachandran, University of Texas at Austin


2:00 PM - 4:00 PM
SESSION 5
Communication and Protocols for Clusters
Chair: Liviu Iftode
Rutgers University

A Consistent History Link Connectivity Protocol
Paul LeMahieu and Jehoshua Bruck, California Institute of Technology

Performance Evaluation of ServerNet Under Self-Similar Traffic
D.R. Avresky and V. Shurbanov, Boston University, R. Horst, Compaq Tandem Labs

Low Latency Message Passing on Workstation Clusters Using SCRAMNet
Vijay Moorthy, Matthew G. Jacunski, Manoj Pillai, Peter P. Ware, Dhabaleswar K. Panda, Thomas W. Page Jr., and P. Sadayappan, Ohio State University, V. Nagarajan and Johns Daniel, SYSTRAN Corporation

Cashmere-VLM: Remote Memory Paging for Software Distributed Shared Memory
Sandhya Dwarkadas, University of Rochester, Nikolaos Hardavellas, Leonidas Kontothanassis, and Rishiyur Nikhil, Compaq Cambridge Research Lab, Robert Stets, University of Rochester

The Computational Co-op: Gathering Clusters Into a Metacomputer
Walfredo Cirne and Keith Marzullo, University of California at San Diego

Reducing System Overheads in Home-Based Software DSMs
Weiwu Hu, Weisong Shi, and Zhimin Tang, Chinese Academy of Sciences, China


2:00 PM - 4:00 PM
SESSION 6
Communication Libraries
Chair: Gul Agha
University of Illinois at Urbana-Champaign

Exploiting Global Structure for Performance on Clusters
Stephen Donaldson and Jonathan M.D. Hill, Oxford University, United Kingdom, David B. Skillicorn, Queen's University, Canada

Implementing Efficient MPI on LAPI for IBM RS/6000 SP Systems: Experiences and Performance Evaluation
Mohammad Banikazemi, Ohio State University, Rama K. Govindaraju and Robert Blackmore, IBM, Dhabaleswar K. Panda, Ohio State University

PM-PVM: A Portable Multithreaded PVM
C.M.P. Santos and J.S. Aude, Federal University of Rio de Janeiro, Brazil

tmPVM -- Task Migratable PVM
C.P. Tan, W.F. Wong, and C.K. Yuen, National University of Singapore, Singapore

A Ubiquitous Message Passing Interface Implementation in Java: jmpi
Kivanc Dincer, Baskent University, Turkey


4:40 PM - 5:00 PM
Industrial Track Presentation
Annapolis Micro Systems, Inc.
Web site: http://www.annapmicro.com

FPGA-Based Architecture for High Speed Serial Processing
Paul Kowalewski and Robert L. Donaldson



5:00 PM - 7:00 PM
SESSION 7
Routing and Broadcasting I
Chair: Frank Dehne
Carleton University, Canada

On-Demand Multicast Routing Scheme and Its Algorithms
Te-Chou Su and Jia-Shung Wang, National Tsing Hua University, Taiwan

Fault-Tolerant Routing Algorithms for Hypercube Networks
Keiichi Kaneko and Hideo Ito, Chiba University, Japan

Dynamic Interval Routing on Asynchronous Rings
Danny Krizanc, Carleton University, Canada, Flaminia L. Luccio, Universita degli Studi di Milano, Italy, Rajeev Raman, King's College London, United Kingdom

Optimally Scaling Permutation Routing on Reconfigurable Linear Arrays with Optical Buses
Jerry L. Trahan and Anu G. Bourgeois, Louisiana State University, Yi Pan, University of Dayton, Ramachandran Vaidyanathan, Louisiana State University


5:00 PM - 7:00 PM
SESSION 8
Miscellaneous Architecture
Chair: Alok Choudhary
Northwestern University

A Comparison of Router Architectures for Virtual Cut-Through and Wormhole Switching in a NOW Environment
J. Duato, A. Robles, and F. Silla, Universidad Politecnica de Valencia, Spain, R. Beivide, Universidad de Cantabria, Spain

Dynamically Scheduling the Trace Produced During Program Execution into VLIW Instructions
Alberto Ferreira de Souza and Peter Rounce, University College London, United Kingdom

Segment Directory Enhancing the Limited Directory Cache Coherence Schemes
Jong Hyuk Choi and Kyu Ho Park, Korea Advanced Institute of Science and Technology, Korea

Shuffle Memory System
Kichul Kim, University of Seoul, Korea

An Efficient Parallel Prefix Sums Architecture with Domino Logic
Rong Lin, SUNY at Geneseo, Koji Nakano, Nagoya Institute of Technology, Japan, Stephan Olariu, Old Dominion University, Albert Y. Zomaya, University of Western Australia, Australia

5:00 PM - 7:00 PM
SESSION 9
Advanced Software for Applications Support
Chair: Ahmed Sameh
Purdue University

The Performance of Coordinated and Independent Checkpointing
Luis M. Silva and Joao Gabriel Silva, Universidade de Coimbra, Portugal

Automatic Array Alignment in Parallel Matlab Scripts
Igor Z. Milosavljevic and Marwan A. Jabri, University of Sydney, Australia

Implementation of NAS Parallel Benchmarks in High Performance Fortran
Michael Frumkin, Haoqiang Jin, and Jerry Yan, MRJ Technology Solutions, Inc.

Parallel Program Archetypes
Berna L. Massingill and K. Mani Chandy, California Institute of Technology

Distributed, Scalable, Dependable Real-Time Systems: Middleware Services and Applications
Lonnie R. Welch and Binoy Ravindran, Virginia Polytechnic Institute and State University, Prashant A. Shirolkar, Shafqat M. Anwar, Terry Sergeant, and Behrooz A. Shirazi, University of Texas at Arlington, Paul Werme, Michael W. Masters, Robert D. Harrison, Wayne Mills, Tuy Do, and Judy Lafrata, Naval Surface Warfare Center, Steve Sharp, George Bilowus, Mark Swick, Jim Hoppel, and Joe Caruso, Computer Sciences Corporation

OpenMP for Networks of SMPs
Y. Charlie Hu, Honghui Lu, Alan L. Cox, and Willy Zwaenepoel, Rice University



COMMERCIAL EXHIBITS