Invited Talks
Important Dates
Program - NEW
Call for Papers
Call for Tutorials
Local Information
São Paulo - Brazil
(Parallel Event)

NEC Awards

Program (NEW)

Monday, November 10, 2003
08:00 - 09:30 Conference Registration
09:00 - 09:30 Coffee Break
09:30 - 11:30 Tutorial 2: Massively Parallel Computing
José Eduardo Moreira, IBM
11:30 - 12:45

Session 4: Grid, Cluster and Pervasive Computing

Performance Issues of Bandwidth Reservations for Grid Computing
Lars Burchard, Hans-Ulrich Heiss, Cesar De Rose

An Evaluation of cJava System Architecture
Anderson Faustino, Marcelo Lobosco, Claudio Amorim

ProGrid: A Proxy-Based Architecture for Grid Operation and Management
Paulo Capellotto Costa, Hélio Guardia, Sérgio Donizetti Zorzo

12:45 - 14:00 Lunch
14:00 - 14:50

Session 2: Processor Architectures

Complex Branch Profiling for Dynamic Conditional Execution
Rafael dos Santos, Tatiana dos Santos, Maurício Pilla, Philippe Navaux, Sergio Bampi, Mario Nemirovsky

The Limits of Speculative Trace Reuse on Deeply Pipelined Processors
Maurício Pilla, Philippe Navaux, Amarildo da Costa, Felipe França, Bruce Childers, Mary Lou Soffa

14:50 - 16:05

Session 1: Cache and Memory Architectures

Exploring Memory Hierarchy with ArchC
Pablo Viana, Edna Barros, Sandro Rigo, Rodolfo Azevedo, Guido Araujo

Adaptive Compressed Caching: Design and Implementation
Rodrigo S. de Castro, Alair Pereira do Lago, Dilma Menezes da Silva

Enabling Dual-Core Mode in BlueGene/L: Challenges and Solutions
George Almasi, Leonardo Bachega, Siddhartha Chatterjee, Manish Gupta, Derek Lieber, Xavier Martorell

16:05 - 16:30 Coffee Break
16:30 - 17:00

Welcoming Remarks
Presence of the Minister of Science & Technology

17:00 - 18:00 Computer Architecture & Technology: The Road Ahead
Michael J. Flynn, University of Stanford, USA

Welcome Reception


Tuesday, November 11, 2003
08:00 - 10:00

Tutorial 3: Grid Computing for Bag-of-Tasks Applications
Walfredo Cirne, Universidade Federal de Campina Grande

10:00 - 10:15 Coffee Break
10:15 - 11:15 From Superscalar and Multiprocessor to Massively Multithreading
Mario D. Nemirovsky, Kayamba, Inc.
11:15 - 12:30

Session 5: High Performance Applications I

Optimizing Packet Capture on Symmetric Multiprocessing Machines
Gianluca Varenni, Mario Baldi, Loris Degioanni, Fulvio Risso

A parallel implementation of the LTSn method for a radiative transfer problem
Roberto Souto, Haroldo Fraga de Campos Velho, Stephan Stephany, Airam Jonatas Preto, Cynthia Feijó Segatto, Marco Túlio Vilhena

Parallel implementation of a lattice-gauge-theory code: studying quark confinement on PC clusters
Attilio Cucchieri, Tereza Mendes, Andre Taurines, Gonzalo Travieso

12:30 - 14:00 Lunch
14:00 - 15:00 Current Trends in SMP Design
Pratap Pattnaik, Senior Manager, Scalable Systems Group, IBM Research Division
15:00 - 16:40

Session 3: Languages and Tools for Parallel and Distributed Programming

JRastro: A Trace Agent for Debugging Multithreaded and Distributed Java Programs
Gabriela Jacques da Silva, Lucas Mello Schnorr, Benhur Stein

On the Implementation of SPMD Applications Using Haskell#
Francisco Carvalho, Rafael Lins, Nivia Quental

Performance Analysis of DECK Collective Communication Service
Rafael Silva, Delcino Picinin, Marcos Barreto, Rafael Avila, Tiarajú Diverio, Philippe Navaux

A Modeling Methodology and Pre-Runtime Scheduling for Embedded Real-Time Software
Raimundo Barreto, Paulo Maciel, Sergio Cavalcante

16:40 - 16:55 Coffee Break
16:55 - 18:10

Session 6: High Performance Applications II

Applying Scheduling by Edge Reversal to Constraint Partitioning
Marluce Pereira, Patrícia Vargas, Felipe França, Maria Clícia Castro, Inês Dutra

Comparison of Genomes using High-Performance Parallel Computing
Nalvo Almeida, Carlos Alves, Edson Cáceres, Siang Song

Finite Difference Simulations of the Navier-Stokes Equations using Parallel Distributed Computing
João Angeli, Andrea Valli, Neyval Reis Jr., Alberto De Souza

20:00 Gala Dinner


Wednesday, November 12, 2003
08:00 - 10:00 SBAC-PAD Meeting
10:00 - 10:15 Coffee Break
10:15 - 11:15 Architectural Requirements for Large Scale Internet Services
Luiz Andre Barroso, Senior Staff Engineer at Google Inc.
11:15 - 12:30

Session 8: Load Balancing and Scheduling

Dynamic Load Balancing in PC Clusters: An Application to a Multi-Physics Model
Ricardo Dorneles, Rogerio Rizzi, Tiarajú Diverio, Philippe Navaux

Hybrid Task Scheduling: Integrating Static and Dynamic Heuristics
Cristina Boeres, Alexandre Lima, Vinod Rebello

Load Balancing on Stateful Clustered Web Servers
George Teodoro, Tulio Tavares, Bruno Coutinho, Wagner Meira, Jr.,
Dorgival Guedes

12:30 - 14:00 Lunch
14:00 - 15:40

Session 7: Parallel and Distributed Algorithms

New Parallel Algorithms for Frequent Itemset Mining in Very Large Databases
Adriano Veloso, Wagner Meira, Jr.

BSP/CGM Algorithm for Maximum Matching in Convex Bipartite Graphs
José Soares, Marco Stefanes

A BSP/CGM Algorithm for Computing Euler Tours in Graphs
Edson Cáceres, Claudia Nasu

Performance Analysis Issues for Parallel Implementations of Propagation Algorithm
Leonardo Brenner, Luiz Gustavo Fernandes, Paulo Fernandes, Afonso Sales

15:40 - 15:55 Coffee Break
15:55 - 17:10

Session 9: Benchmarking, Performance Measurements and Analysis

Profiling and Optimization of Software-Based Network-Analysis Applications
Loris Degioanni, Mario Baldi, Fulvio Risso, Gianluca Varenni

PM2P: A tool for performance monitoring of message passing applications in COTS PC clusters
Maya Haridasan, Gerson Pfitscher

Boosting Performance for I/O-Intensive Workload by Preemptive Job Migrations in a Cluster System
Xiao Qin, Hong Jiang, Yifeng Zhu, David Swanson

Session 10: Reconfigurable Systems

Three Hardware Implementations for the Binary Modular Exponentiation: Sequential, Parallel and Systolic
Nadia Nedjah, Luiza Mourelle

Fast Parallel FFT on A Reconfigurable Computation Platform
Amir Kamalizad, Chengzhi Pan, Nader Bagherzadeh

X4CP32: A New Hybrid Parallel/Reconfigurable General-Purpose Processor
Rodrigo Soares, Arnaldo Azevedo, Ivan Silva




17:15 Closing Remarks

© Copyright - All rights reserved