The Community for Technology Leaders
Computer Architecture and High Performance Computing, Symposium on (2004)
Foz do Igua?u, PR - Brazil
Oct. 27, 2004 to Oct. 29, 2004
ISSN: 1550-6533
ISBN: 0-7695-2240-8
TABLE OF CONTENTS

Cache filtering techniques to reduce the negative impact of useless speculative memory references on processor performance (PDF)

O. Mutlu , Dept. of Electr. & Comput. Eng., Texas Univ., Austin, TX, USA
H. Kim , Dept. of Electr. & Comput. Eng., Texas Univ., Austin, TX, USA
D.N. Armstrong , Dept. of Electr. & Comput. Eng., Texas Univ., Austin, TX, USA
Y.N. Patt , Dept. of Electr. & Comput. Eng., Texas Univ., Austin, TX, USA
pp. 2-9

Self-monitored adaptive cache warm-up for microprocessor simulation (PDF)

Y. Luo , Texas Univ., Austin, TX, USA
L.K. John , Texas Univ., Austin, TX, USA
pp. 10-17

The eDRAM based L3-cache of the BlueGene/L supercomputer processor node (PDF)

M. Ohmacht , IBM Thomas J. Watson Res. Center, Hawthorne, NY, USA
D. Hoenicke , IBM Thomas J. Watson Res. Center, Hawthorne, NY, USA
R. Haring , IBM Thomas J. Watson Res. Center, Hawthorne, NY, USA
A. Gara , IBM Thomas J. Watson Res. Center, Hawthorne, NY, USA
pp. 18-22

A study of errant pipeline flushes caused by value misspeculation (PDF)

D. Balkan , Dept. of Comput. Sci., SUNY, Binghamton, NY, USA
pp. 32-39

Design space exploration using T&D-Bench (PDF)

S.N. Soares , PGCC, Univ. Fed. do Rio Grande do Sul, Brazil
pp. 40-47

Value predictors for reuse through speculation on traces (PDF)

M.L. Pilla , Comput. Sci. Inst., Univ. Fed. do Rio Grande do Sul, Porto Alegre, Brazil
P.O.A. Navaux , Comput. Sci. Inst., Univ. Fed. do Rio Grande do Sul, Porto Alegre, Brazil
pp. 48-55

Program Committee (PDF)

pp. xi-xii

Reviewers (PDF)

pp. xiii
Session 1: Cache and Memory Architectures

Cache Filtering Techniques to Reduce the Negative Impact of Useless Speculative Memory References on Processor Performance (Abstract)

Onur Mutlu , The University of Texas at Austin
Hyesoon Kim , The University of Texas at Austin
David N. Armstrong , The University of Texas at Austin
Yale N. Patt , The University of Texas at Austin
pp. 2-9

Self-Monitored Adaptive Cache Warm-Up for Microprocessor Simulation (Abstract)

Yue Luo , University of Texas at Austin, USA
Lizy K. John , University of Texas at Austin, USA
Lieven Eeckhout , Ghent University, Belgium
pp. 10-17

The eDRAM based L3-Cache of the BlueGene/L Supercomputer Processor Node (Abstract)

Martin Ohmacht , IBM T. J. Watson Research Center
Dirk Hoenicke , IBM T. J. Watson Research Center
Ruud Haring , IBM T. J. Watson Research Center
Alan Gara , IBM T. J. Watson Research Center
pp. 18-22

Multi-Profile Instruction Based Compression (Abstract)

Eduardo Wanderley Netto , CEFETRN / IC-UNICAMP, Brazil
Rodolfo Azevedo , IC-UNICAMP, Brazil
Paulo Centoducatte , IC-UNICAMP, Brazil
Guido Araujo , IC-UNICAMP, Brazil
pp. 23-29
Session 2: Processor Architectures I

A Study of Errant Pipeline Flushes Caused by Value Misspeculation (Abstract)

Deniz Balkan , SUNY, Binghamton, NY
John Kalamatianos , AMD, Inc., Boxborough, MA
David Kaeli , Northeastern University, Boston, MA
pp. 32-39

Design Space Exploration using T&D-Bench (Abstract)

Sandro Neves Soares , PGCC - UFRGS / UCS - CARVI
Fl?vio Rech Wagner , Institute of Informatics - UFRGS
pp. 40-47

Value Predictors for Reuse through Speculation on Traces (Abstract)

Maur?cio L. Pilla , UFRGS - Brazil
Philippe O. A. Navaux , UFRGS - Brazil
Bruce R. Childers , Univ. of Pittsburgh - USA
Amarildo T. da Costa , IME - Brazil
Felipe M. G. Fran? , COPPE/UFRJ - Brazil
pp. 48-55
Session 3: Processor Architectures II

IATO: A Flexible EPIC Simulation Environment (Abstract)

Amaury Darsch , IRISA/INRIA, France
Andr? Seznec , IRISA/INRIA, France
pp. 58-65

ArchC: A SystemC-Based Architecture Description Language (Abstract)

Sandro Rigo , University of Campinas, Brazil
Guido Ara? , University of Campinas, Brazil
Marcus Bartholomeu , University of Campinas, Brazil
Rodolfo Azevedo , University of Campinas, Brazil
pp. 66-73

Optimizations for Compiled Simulation Using Instruction Type Information (Abstract)

Marcus Bartholomeu , University of Campinas, Brazil
Rodolfo Azevedo , University of Campinas, Brazil
Sandro Rigo , University of Campinas, Brazil
Guido Araujo , University of Campinas, Brazil
pp. 74-81
Session 4: Languages and Tools for Parallel and Distributed Programming

Improving Server Performance on Transaction Processing Workloads by Enhanced Data Placement (Abstract)

Juan Rubio , The University of Texas at Austin
Charles Lefurgy , IBM Austin Research Lab, Austin, TX
Lizy K. John , The University of Texas at Austin
pp. 84-91

High Performance Communication System Based on Generic Programming (Abstract)

Andr? Lu?s Gobbi Sanches , Universidade Federal de Santa Catarina - UFSC
Fernando Roberto Secco , Universidade Federal de Santa Catarina - UFSC
Ant?nio Augusto Fr?hlich , Universidade Federal de Santa Catarina - UFSC
pp. 92-99

Performance Evaluation of a Prototype Distributed NFS Server (Abstract)

Rafael B. ?vila , Instituto de Inform?tica/UFRGS, Brazil
Philippe O. A. Navaux , Instituto de Inform?tica/UFRGS, Brazil
Pierre Lombard , Laboratoire ID/IMAG, France
Adrien Lebre , Laboratoire ID/IMAG, France
Yves Denneulin , Laboratoire ID/IMAG, France
pp. 100-105
Session 5: Grid, Cluster and Pervasive

FlowCert : Probabilistic Certification for Peer-to-Peer Computations (Abstract)

S?bastien Varrette , Laboratoire ID-IMAG, France
Jean-Louis Roch , Laboratoire ID-IMAG, France
Franck Lepr?vost , Universit? du Luxembourg, Luxembourg
pp. 108-115

A Performance Evaluation of a Quorum-Based State-Machine Replication Algorithm For Computing Grids (Abstract)

Jean-Michel Busca , Universit? Paris 6 - CNRS, France; INRIA Rocquencourt, France
Marin Bertier , Universit? Paris 6 - CNRS, France
Fatima Belkouch , Universit? Lille 2, France
Pierre Sens , Universit? Paris 6 - CNRS, France; INRIA Rocquencourt, France
Luciana Arantes , Universit? Paris 6 - CNRS, France
pp. 116-123

Scheduling in Bag-of-Task Grids: The PAU? Case (Abstract)

Walfredo Cirne , Universidade Federal de Campina Grande
Francisco Brasileiro , Universidade Federal de Campina Grande
Lauro Costa , Universidade Federal de Campina Grande
Daniel Paranhos , Universidade Federal de Campina Grande
Elizeu Santos-Neto , Universidade Federal de Campina Grande
Nazareno Andrade , Universidade Federal de Campina Grande
Miranda Mowbray , Hewlett Packard
Roque Scheer , Hewlett Packard
Jo?o Jornada , Hewlett Packard
pp. 124-131

meμ: Unifying Application Modeling and Cluster Exploitation (Abstract)

Albano Alves , ESTiG-IPB
Ant?nio Pina , DI-UMinho
Jos? Exposto , ESTiG-IPB
Jos? Rufino , ESTiG-IPB
pp. 132-139
Session 6: High Performance Applications

Parallel Implementation of a Lagrangian Stochastic Model for Pollution Dispersion (Abstract)

Debora R. Roberti , Federal University of Santa Maria (UFSM), Santa Maria (RS), Brazil
Roberto P. Souto , National Institute for Space Research (INPE), S?o Jos? dos Campos (SP), Brazil
Haroldo F. de Campos Velho , National Institute for Space Research (INPE), S?o Jos? dos Campos (SP), Brazil
Gervasio A. Degrazia , Federal University of Santa Maria (UFSM), Santa Maria (RS), Brazil
Domenico Anfossi , Italian National Research Council (CNR)
pp. 142-149

A Parallel Engine for Graphical Interactive Molecular Dynamics Simulations (Abstract)

Eduardo Rocha Rodrigues , Brazilian Institute for Space Research (INPE)
Airam Jonatas Preto , Brazilian Institute for Space Research (INPE)
Stephan Stephany , Brazilian Institute for Space Research (INPE)
pp. 150-157

Parallel Adaptive Mesh Coarsening for Seismic Tomography (Abstract)

Marc Grunberg , IPGS, CNRS-ULP, Strasbourg
St?phane Genaud , LSIIT-ICPS, CNRS-ULP, Illkirch
Catherine Mongenet , LSIIT-ICPS, CNRS-ULP, Illkirch
pp. 158-165
Session 7: Parallel and Distributed Algorithms

Revisiting a BSP/CGM Transitive Closure Algorithm (Abstract)

Edson Norberto C?ceres , Federal University of Mato Grosso do Sul, Brazil
Cristiano Costa Argemom Vieira , Federal University of Mato Grosso do Sul, Brazil
pp. 174-179

Improving Parallel Execution Time of Sorting on Heterogeneous Clusters (Abstract)

Christophe C?rin , Universit? de Picardie Jules Verne, France
Michel Koskas , Universit? de Picardie Jules Verne, France
Hazem Fkaier , ?cole Sup?rieure des Sciences et Techniques de Tunis, Tunisie
Mohamed Jemni , ?cole Sup?rieure des Sciences et Techniques de Tunis, Tunisie
pp. 180-187

An Approach for Pre Runtime Scheduling in Embedded Hard Real Time Systems with Power Constraints (Abstract)

Eduardo Tavares , Federal University of Pernambuco (UFPE)
Raimundo Barreto , Federal University of Amazonas (UFAM)
Meuse Oliveira J?nior , Federal University of Pernambuco (UFPE)
Paulo Maciel , Federal University of Pernambuco (UFPE)
Mar?lia Neves , Federal University of Pernambuco (UFPE)
Ricardo Lima , Pernambuco State University
pp. 188-195
Session 8: Load Balancing and Scheduling

Graph Partitioning with the Party Library: Helpful-Sets in Practice (Abstract)

Burkhard Monien , Universit?t Paderborn
Stefan Schamberger , Universit?t Paderborn
pp. 198-205

On the Combined Scheduling of Malleable and Rigid Jobs (Abstract)

Jan Hungersh?fer , Paderborn Center for Parallel Computing, Germany
pp. 206-213

A Cluster-based Strategy for Scheduling Task on Heterogeneous Processors (Abstract)

Cristina Boeres , Universidade Federal Fluminense (UFF), Brazil
Jos? Viterbo Filho , Universidade Federal Fluminense (UFF), Brazil
Vinod E. F. Rebello , Universidade Federal Fluminense (UFF), Brazil
pp. 214-221
Session 9: Benchmarking, Performance Measurements and Analysis

Characterizing the Dynamic Behavior of Workload Execution in SVM systems (Abstract)

Salvador Petit , Universidad Polit?cnica de Valencia, Spain
Julio Sahuquillo , Universidad Polit?cnica de Valencia, Spain
Ana Pont , Universidad Polit?cnica de Valencia, Spain
David Kaeli , Northeastern University, Boston, Massachusetts
pp. 230-237

A Performance Evaluation of ARM ISA Extension for Elliptic Curve Cryptography over Binary Finite Fields (Abstract)

Sandro Bartolini , University of Siena, Italy
Irina Branovic , University of Siena, Italy
Roberto Giorgi , University of Siena, Italy
Enrico Martinelli , University of Siena, Italy
pp. 238-245

PEMPIs: A New Methodology for Modeling and Prediction of MPI Programs Performance (Abstract)

Edson T. Midorikawa , Polytechnic School, University of S?o Paulo, Brazil
Helio M. de Oliveira , Polytechnic School, University of S?o Paulo, Brazil
Jean M. Laine , Polytechnic School, University of S?o Paulo, Brazil
pp. 246-253

Performance Characterisation of Intra-Cluster Collective Communications (Abstract)

Luiz Angelo Barchet-Estefanel , ID - IMAG Laboratory, France
Gr?gory Mouni? , ID - IMAG Laboratory, France
pp. 254-261

Author Index (PDF)

pp. 263-264
89 ms
(Ver 3.3 (11022016))