This is an archived page of the 2002 conference

program info

The Third LCI Conference took place at the Hilton St. Petersburg in St. Petersburg, Florida, and it featured speakers from academia, government research labs, and industry involved in Linux high-performance computing. These speakers addressed efforts to integrate and develop Linux clusters for high-performance computing, and efforts to develop science and engineering applications for Linux clusters. The keynote speakers for the conference were Dr. Mike Turff, Global Manager DP Systems, PGS Data Processing; Dr. Daniel Reed, Director of the National Center for Supercomputing Applications (NCSA); and Dr. William Pulleyblank, Director of Exploratory Server Systems in IBM's Research Division and the Director of the IBM Deep Computing Institute.

wednesday | thursday | friday


Wednesday, October 23, 2002

8:45am Welcome to the Conference
9:00am Plenary Session I:
Seismic & Linux: From Curiosity to Commodity
Dr. Mike Turff, Global Manager DP Systems, PGS Data Processing
10:15am Break
10:30am Cluster Busters Scalability and Performance of Salinas on the Computational Plant
Ron Brightwell, Univ of New Mexico, USA

Real Time Response to Streaming Data on Linux Clusters
George Turner, Indiana Univ, USA
12:00pm Lunch 
1:30pm Applications I:  Applications Performance Analysis

Scaling Behavior of Linear Solvers on Large Linux Clusters
John Fettig, NCSA/Univ of Illinois, USA

Benchmarking of a Commercial Linux Cluster
Daniele Tessera, Università di Pavia, Italy
Systems I:  System Performance Issues

Checkpointing and Migration of Parallel Processes Based on Message Passing Interface
Zhang Youhui, Tsinghua University, China

An Analytical Model to Evaluate the Performance of Cluster Architectures
César A. F. De Rose, PUCRS - Faculdade de Informatica, Brazil
3:00pm Break
3:30pm Vendor Presentations I:

Delivering Linux Cluster Solutions to the High-Performance Marketplace
Joseph Banas, Go to Market Leader, IBM eServer Linux Cluster

David Addison, Quadrics
5:00pm End of Presentations
6:00-
9:00pm
Conference Reception
   

Thursday, October 24, 2002

9:00am Plenary Session II:
Scaling Up and Scaling Down
Dr. Daniel A. Reed, Director of the National Center for Supercomputing Applications (NCSA)
10:15am Break
10:30am Programming Models

Remote Memory Operations on Linux Clusters Using the Global Arrays Toolkit, GPSHMEM and ARMCI
Chona S. Guiang, TACC/Univ of Texas, USA

Supercomputing in Plain English: Teaching High Performance Computing to Inexperienced Programmers
Henry Neeman, Oklahoma Univ, USA
12:00pm Lunch
1:30pm Applications II: Applications Performance on Clusters

A Comparative Study of the Performance of a CFD program across different Linux Cluster Architectures
Thomas Hauser, Univ of Kentucky, USA

Scalability of a Tera-scale Linux-based Clusters from Parallel Ab Initio Molecular Dynamics
Jungwoo Hong, KISTI, Korea

openMosix vs Beowulf: a Case Study
Stefano Cozzini, INFM, Italy
Systems II:  Cluster Management and Monitoring

High-Performance Linux Cluster Monitoring Using Java
David Henry, Linux NetworX

CRONO: A Configurable Management System for Linux Clusters
César A. F. De Rose, PUCRS - Faculdade de Informatica, Brazil

Parallel, Distributed Scripting with Python
Patrick Miller, Lawrence Livermore National Laboratory, USA

 

3:45pm Break  
4:00pm Sponsor Presentations II:

Designing Supercomputers that Fit Your Needs
Tim Mattson
4:45pm End of Presentations
   

Friday, October 25, 2002

9:00am Plenary Session III:  

Challenges of Terascale Computing
Dr. William Pulleyblank, Director of Exploratory Server Systems in IBM's Research Division and Director of the IBM Deep Computing Institute
10:15am Break 
10:30am Applications III: Experiences with Portability and I/O

Adventures with Portability
Elizabeth Post, Lincoln Univ, New Zealand

Experiences with the Parallel Virtual File System (PVFS) in Linux Clusters
Kent Milfeld, TACC/Univ of Texas, USA
Systems III: Processor and Network Performance

A Study of Hyper-Threading in High Performance Computing Clusters
Tau Leng, Dell 

A Performance Comparison of Myrinet Protocol Stacks
Ron Brightwell, Univ of New Mexico, USA
12:00pm Lunch  
1:30pm Panel Discussion:
Users, Applications, and Systems: Linux - HPC in 2005
Moderator: Bob Balance, HPCERC, University of New Mexico
Panelists: Frank Gilfeather, HPCERC, University of New Mexico; Tim Mattson, Intel; William Pulleyblank, IBM; and Neil Pundit, Sandia
3:00pm Adjourn Conference