October 27-29, 2003
Symposium Site: 
Eldorado Hotel, 309 West San Francisco Street
Sante Fe, New Mexico

• Proceedings | Download PDF
• Symposium Archive (previous LACSI Symposiums)

Agenda for LACSI Symposium 2003

Workshops | Poster Sessions | Papers | Panels

Monday, October 27: Workshop Sessions

All Day Workshops

Title: Clustermatic: A Revolutionary Approach to Simplified Cluster Computing
Organizers: Sung-Eun Choi, Ronald Minnich, Matthew Sottile, Gregory Watson (LANL)

Title: System and Application Performance
Organizers: Adolfy Hoisie (Performance and Architecture Lab (PAL), (LANL))
Allen Malony (University of Oregon)
John Mellor-Crummey (Rice University)

Morning Workshops: 9:00am - 12:30pm

Title: Interoperability of Object Libraries for Scientific Computing
Organizers: Anthony D. Padula, William W. Symes (Rice University)
Roscoe A. Bartlett (Sandia National Laboratory)
Craig Rasmussen (LANL)

Title: Mimetic Methods for Radiation Transport and Diffusion
Organizers: Jim Morel (CCS-4, LANL)
Yuri Kuznetsov (University of Houston)
Mikhail Shashkov (T-7, LANL)

Title: High Availability and Performance Computing
Organizers: Stephen L. Scott (Oak Ridge National Laboratory)
Chokchai Leangsuksun (Louisiana Tech University)

Afternoon Workshops: 2:00pm - 5:30pm

Title: Simulation Driven Optimization
Organizers: Matthias Heinkenschloss, William W. Symes, Yin Zhang (Rice University)

Title: Particle Simulation: Challenges and Prospects
Organizers: Salman Habib (LANL)
Robert D. Ryne (LBNL)

Title: A Tutorial on Adaptive MPI and Charm++
Organizers: Laxmikant V. Kale, Orion Lawlor (University of Illinois at Urbana-Champaign)

• 6:00pm – 8:00pm: Welcoming Reception and Poster Presentations

Workshops/tutorials on subjects of special interest to attendees, and the Welcome Reception and Poster Exhibit were featured. The posters were also made available over the next two days for additional inspection.

Poster Presentations

• Connection Management for Massively Scalable TCP
  Patricia Eileen Crowley, Arthur B. Maccabe, and Breanne A. Duncan (The University of New Mexico)
• Grid Performance Prediction with Performance Skeletons
  Sukhdeep Sodhi and Jaspal Subhlok (University of Houston)
• A Portable High Performance Co-Array Fortran Compiler
  Yuri Dotsenko, Cristian Coarfa, and John Mellor-Crummey (Rice University)
• An In-House Development for Visualization in Molecular Science
  Jean M. Favre and Mario Valle (Swiss Center for Scientific Computing)
• Dragon: Program Analysis and Development Support for Open MP Codes
  Barbara Chapman, Oscar Hernandez, Lei Huang, Yi Wen, Tien-hsiung Weng, and Zhenying Liu (University of Houston)
• A Sample-Driven Call Stack Profiler
  Nathan Froyd, John Mellor-Crummey, Nathan Tallent (Rice University)
• Clam: Light Communication Layer for Asynchronous Mobile Computations
  Andriy Fedorov and Nikos Chrisochoides (College of William and Mary)
• PAPI Version 3
  Jack Dongarra, Kevin London, Shirley Moore, Philip Mucci, Daniel Terpstra, Haihang You, and Min Zhou (University of Tennessee)
• Processor-Level Partitioning of Kernel and User Mode Functionality on SMP Systems
  Carl A. Sylvia (The University of New Mexico)
• Model Coupling Toolkit Benchmarks
  J. Walter Larson, Robert L. Jacob, and Everest T. Ong (Argonne National Laboratory)
• Fault Injection into MPI Programs
  Charng-da Lu, Karthik Pattabiraman, and Daniel A. Reed (University of Illinois at Urbana-Champaign)
• High Performance Object-Oriented Java Programming: Fiction or Reality?
  Zoran Budimlic (Rice University), Giovanni Lapenta (LANL), Stefano Markidis (Politecnico di Torino), and Brian VenderHeyden (LANL)
• Profile-Based Dynamic Voltage Scaling for I/O-Intensive Codes
  Karthik Pattabiraman (LANL and University of Illinois at Urbana-Champaign), Wu-Chun Feng (LANL), and Daniel Reed (University of Illinois at Urbana-Champaign)
• A High-Level Approach to the Synthesis of High-Performance Codes for Quantum Chemistry
  David E Bernholdt, Venkatesh Choppella, and Robert J. Harrison (Oak Ridge National Laboratory); Alexander Auer and Marcel Nooijen (University of Waterloo); Gerald Baumgartner, Alina Bibireata, Daniel Cociorva, Xioyang Gao, Sriram Krishnamoorthy, Sandhya Krishnan, Chi-Chung Lam, Qingda Lu, Russell M. Pitzer, P. Sadayappan, and Alexander Sibiryakov (Ohio State University); So Hirata (Pacific Northwest National Laboratory); J. Ramanujam (Louisiana State University)
• Optimizing Collective Multicast on PSC Lemieux
  Sameer Kumar and Laxmikant V. Kale (University of Illinois at Urbana-Champaign)
• Supporting Generic Programming in a Multi-Language Component-Based Environment
  Wael R. Elwasif, Thomas C. Schulthess, and David E. Berhnoldt (Oak Ridge National Laboratory); Gregory Brown (Florida State University)
• Design and Implementation of the Prophesy Automated Model Builder for Parallel and Grid Applications
  Joseph Paris (Northwestern University), Xingfu Wu and Valerie Taylor (Texas A&M University)

Tuesday, October 28:

• 9:00am Welcoming remarks
• 9:10am: Keynote Address – “Challenges in Cyber Trust”
  Dr. Gene Spafford, Purdue University
  Center for Education and Research in Information Assurance and Security

Abstract:
The last 30+ years of computing have primarily been focused on getting computing to work, at least most of the time. Now we are employing computing equipment in everything from wristwatches to the power grid, from automobiles to air traffic control. It is a vital part of every critical infrastructure to society, as well as most of our research and entertainment industries. Thus, we now need to focus on making the computing reliable and trustworthy. The goal of computing faster may need to be tempered by the question (from another field): Unsafe at any speed? What are some of the ways in which large-scale computing can assist in this effort? Are the security challenges posed by grid and cluster computing unique? This talk will explore some ideas from this space, and encourage the audience to start thinking of their own examples.

• 10:00am Break
• 10:30am – Reviewed Papers I: Partitioning, Distribution, and Scheduling
  • Parallel Remote Method Invocation and M-by-N Data Redistribution
    Kostadin Damevski and Steven Parker (University of Utah)
  • An Approach to Parallel MxN Communication
    Felipe Bertrand, Yongquan Yuan, Kenneth Chiu, and Randall Bramley (Indiana University)
  • A Partitioner-Centric Model for SAMR Partitioning Trade-off Optimization: Part I
    Johan Steensland and Jaideep Ray (Sandia National Laboratories)
  • Opportune Job Shredding: An Effective Approach for Scheduling Parameter Sweep Applications
    Rohan Kurian, Pavan Balaji, and P. Sadayappan (The Ohio State University)
• 12:30pm – 2:00pm Lunch
• 2:30pm – Reviewed Papers II: Compilation and Memory Hierarchies
  • Investigating Adaptive Compilation Using the MIPSpro Compiler
    Keith D. Cooper and Todd Waterman (Rice University)
  • On Reducing Storage Requirement of Scientific Applications
    Guohua Jin and John Mellor-Crummey (Rice University)
  • Regression-Based Multi-Model Prediction of Data Reuse Signature
    Xipeng Shen, Yutao Zhong, and Chen Ding (University of Rochester)
• 3:30pm – 4:00pm Break
• 4:00pm Reviewed Papers III: Cluster Communication and Storage
  • Implementing Scalable Diskless Clusters Using NFS
    James Howard Laros III and Lee Ward (Sandia National Laboratories)
  • The LAM/MPI Checkpoint/Restart Framework: System-Initiated Checkpointing
    Sriram Sankaran, Jeffrey M. Squyres, Brian Barrett, Andrew Lumsdaine, Jason Duell, Paul Hargrove, and Eric Roman (Indiana University and Lawrence Berkeley National Laboratory)
  • Fault Tolerant Communication Library and Applications for High Performance Computing
    Graham Edward Fagg, Edgar Gabriel, Zizhon Chen, Thara Angskun, George Bosilica, Antonin Bukovsky, and Jack Dongarra (ICL/University of Tennessee, Knoxville)
  • High Performance Broadcast Support in LA-MPI over Quadrics
    Weikuan Yu, Sayantan Sur, and D.K. Panda (The Ohio State University); Rob T. Aulwes and Rich L. Graham (LANL)

Wednesday, October 29: Panel Discussions

Four panel discussions took place focusing on important issues for researchers and managers in high performance computing and a Closing Reception ended the LACSI Symposium.

• 8:00am Breakfast
• 9:00am – Panel I
  • Panel I: The Earth Simulator: One Year Later
    Discussants will review aspects of hardware, software, and applications of the system that has caused much controversy when first introduced in 2002.
• 10:30am Break
• 11:00am – Panel II
  • Panel II: Experiences with TOP500 Linux Systems
    Systems running Linux have made steady progress in invading the high end of the TOP500 list. Representatives from organizations with the most powerful systems will discuss their experiences.
• 12:30 Lunch
• 2:00pm – Panel III
  • Panel III: High Productivity Computer Systems, Phase II
    Representatives from Cray, IBM, and Sun will discuss their efforts in Phase II of DARPA’s HPCS program.
• 3:30pm Break
• 4:00PM – 5:30pm Panel IV
  • Panel IV: Federal Programs in High-Performance Computing
    Panelists from Federal agencies that support HPC and that have participated in HECRTF will present per-spectives on the future of Federal programs that support the research, development, and use of high performance computing platforms.
• 5:30pm – 7:00pm Closing Reception