SBAC-PAD 2002 – Program

MONDAY, October 28, 2002

08:00 – 10:00

Conference Registration

10:00 – 10:30

Coffee

10:30 – 11:00

Welcoming Remarks

11:00 – 12:00

Keynote Address: Jack Dongarra – University of Tennessee (USA)
- High Performance Computing, Computational Grid, and Numerical Libraries -

12:00 – 14:00

Lunch

14:00 – 15:00

Keynote Address: Graham F. Carey – the University of Texas at Austin (USA)
- Advances and Challenges in Finite Element Methods and Algorithms for High Performance Computation -

15:00 – 16:15

Session 1: Application-specific Architectures

Instruction Set Extension for Long Integer Modulo Arithmetic on RISC-based Smart Cards
Johann Groβschädl (Austria)

Interactive Ray Tracing Using a SIMD Reconfigurable Architecture
Manuel L. Anido (Brazil), Nozar Tabrizi (USA), Haitao Du (USA), Marcos Sanchez-Elez M. (Spain) and Nader Bagherzadeh (USA)

Architecture of Oscillatory Neural Network for Image Segmentation
Dênis Fernandes (Brazil), Jeferson Polidoro Stedile (Brazil) and Philippe Olivier Alexandre Navaux (Brazil)

16:15 – 16:35

Coffee-Break

16:35 – 17:50

Session 2: High-Performance Applications

Simulating Semiconductor Spectra Emissions in a PC Cluster
Guilherme Matos Sipahi (Brazil) and Gonzalo Travieso (Brazil)

Improving the Communication Network of a Cluster of PC’s to Solve Implicit CFD Problems
Renato S. Silva (Brazil), Marconi F. P. Rivello (Brazil) and Ricardo Amorim (Brazil)

Parallel Boundary Elements Using Lapack and ScaLapack
Manoel T. F. Cunha (Brazil), José C. F. Telles (Brazil) and Alvaro L. G. A. Coutinho (Brazil)

17:50 – 18:00

Short-Break

18:00 – 18:30

Industrial Presentation:
 

 

19:00 –

Welcoming RECEPTION

 

 

 

 

 

TUESDAY, October 29, 2002

08:00 – 09:40

Session 3: Parallel and Distributed Programming: Algorithms and Tools

Efficient Cyclic Weighted Reference Counting
Rafael Dueire Lins (Brazil)

Generating Java Code for TINA Systems
Rafael Paoliello Guimarães (Brazil) and Walter da Cunha Borelli (Brazil)

A Parallel Approximation Hitting Set Algorithm for Gene Expression Analysis
Danielle Passos Ruchkys (Brazil) and Siang W. Song (Brazil)

Implementing Declarative Parallel Bottom-Avoiding Choice
André Rauber Du Bois (UK), Robert Pointon (UK), Hans-Wolfgang Loidl (UK) and Phil Trinder (UK)

09:40 – 10:00

Coffee-Break

10:00 – 11:00

Tutorial 1 – part 1

An Introduction to the Development of Performance Evaluation Tools for Parallel Programming
Luiz DeRose – IBM Research (USA)

Tutorial 2 – part 1

Energy Management for Server Clusters
Ram Rajamony – IBM Austin Research Laboratory (USA)
Ricardo Bianchini – Rutgers University (USA )

11:00 – 12:00

Keynote Address: Mateo Valero – Universitat Politècnica de Catalunya (Spain)
- Will Vector ISAs Survive in the Future? -

12:00 – 14:00

Lunch

14:00 – 15:15

Session 4: Cache and Memory Architectures

An Advanced Filtering TLB for Low Power Consumption
Jin-Hyuck Choi (Korea), Jung-Hoon Lee (Korea), Gi-Ho Park (Korea) and Shin-Dug Kim (Korea)

Minimally-Skewed-Associative Caches
Andrés Djordjalian (Argentina)

Simulating L3 Caches in Real Time using Hardware Accelerated Cache Simulation (HACS): A Case Study with SPECint 2000
Myles Watson (USA) and J. Kelly Flanagan (USA)

15:15 – 16:15

Tutorial 1 – part 2

An Introduction to the Development of Performance Evaluation Tools for Parallel Programming
Luiz DeRose – IBM Research (USA)

Tutorial 2 – part 2

Energy Management for Server Clusters
Ram Rajamony – IBM Austin Research Laboratory (USA)
Ricardo Bianchini – Rutgers University (USA )

16:15 – 16:35

Coffee-Break

16:35 – 18:05

Session 5: Grid, Multimedia, and Cluster-based Computing

GloVE: A Distributed Environment for Low Cost Scalable VoD Systems
Leonardo B. Pinho (Brazil), Edison Ishikawa (Brazil) and Claudio L. Amorim (Brazil)

Multilevel Colaborative Adaptation in High Heterogeneous Distributed Processing
Adenauer Yamin, Iara Augustin, Jorge Barbosa, Luciano Silva, Rodrigo Real, Gerson Cavalheiro, Claudio Geyer (Brazil)

Cluster-based Static Scheduling: Theory and Practice
Cristina Boeres (Brazil) and Vinod E. F. Rebello (Brazil)

The Virtual Cluster: a Dynamic Environment for Exploitation of Idle Network Resources
C. De Rose (Brazil), F. Blanco (Brazil), N. Maillard (Brazil), K. Saikoski (Brazil), R. Novaes (Brazil), O. Richard (France), B. Richard (France)

 

20:00 –

gala dinner

 

 

 

 

WEDNESDAY, October 30, 2002

08:00 – 08:50

Session 6: Distributed Shared Memory

Design and Evaluation of Data Access Prediction Strategies in SDSM Systems
Elcio José Pineschi (Brazil), Claudio Luis Amorim (Brazil) and Maria Clicia Stelling de Castro (Brazil)

Distributed Shared Memory in Kernel Mode
Thobias S. Trevisan (Brazil), Lauro Whately (Brazil), Vítor S. Costa (Brazil) and Claudio L. Amorim (Brazil)

08:50 – 09:40

Academic/Industrial Presentation: LSI-EPUSP / ITUTEC / FINEP
The Design of a Beowulf Cluster: Architectural and Managerial Aspects
Rogério Luiz Iope and Marilda Ferrari Mendes Giafarov

09:40 – 10:00

Coffee-Break

10:00 – 11:00

Tutorial 3 – part 1

Topics in Parallel Algorithms using CGM/MPI
Edson Norberto Cáceres, Henrique Mongelli – Universidade Federal de Mato Grosso do Sul (Brazil)
Siang Wun Song – Universidade Estadual de São Paulo (Brazil)

 

11:00 – 12:00

Keynote Address: Liviu Iftode, Rutgers - University of Maryland (USA)
- Playing Distributed Systems with Memory-to-Memory Communication -

12:00 – 14:00

Lunch

14:00 – 15:15

Session 7: Processor Architecture

Instruction Usage and the Memory Gap Problem
Edil S. T. Fernandes (Brazil), Valmir C. Barbosa (Brazil) and Fabiano Ramos (Brazil)

T&D-Bench: an Environment for Modeling and Simulating Complex Processor Architectures
Douglas M. Becker (Brazil), João Cláudio Otero (Brazil) and Flávio Rech Wagner (Brazil)

Exploiting Loop-Level Parallelism with the Shift Architecture
Clecio Donizete Lima (Japan) and Tadao Nakamura (Japan)

15:15 – 16:15

Tutorial 3 – part 2

Topics in Parallel Algorithms using CGM/MPI
Edson Norberto Cáceres, Henrique Mongelli – Universidade Federal de Mato Grosso do Sul (Brazil)
Siang Wun Song – Universidade Estadual de São Paulo (Brazil)

 

16:15 – 16:35

Coffee-Break

16:35 – 17:50

Session 8: Performance Analysis and Tools

An Evaluation of Processor Co-Allocation for Different System Configurations and Job Structures
Anca I. D. Bucur (The Netherlands) and Dick H. J. Epema (The Netherlands)

Performance Analysis of Parallel Programs using Prober as a Single Aid Tool
Luís Fabrício W. Góes (Brazil), Luiz Eduardo S. Ramos (Brazil) and Carlos Augusto P. S. Martins (Brazil)

Performance and Scalability Measurement of Enterprise JavaBeans Technology
Jenny Yan Liu (Australia)

17:50 – 18:30

Break

18:30 – 18:45

CLosing remarks