High Performance Computing and Simulation Symposium
(HPCS 2008)
April 14-16, 2008
part of the
SCS Spring Simulation Multiconference (SpringSim'08)
A joint symposium of DEVS Integrative M&S (DEVS) and
High Performance Computing (HPC)
|
GENERAL CHAIR |
|
Masha Sosonkina, Ames Laboratory |
|
GENERAL VICE-CHAIR |
|
Tomasz Haupt, Mississippi State University |
|
PROGRAM CHAIR |
|
Cliff Shaffer, Virginia Tech |
|
PROGRAM VICE-CHAIR |
|
Adrian Sandu, Virginia Tech |
|
CALL FOR PAPERS |
|
Download in TEXT format |
|
IMPORTANT DATES |
|
Paper submission due: November 30, 2007 Acceptance Notification: January 7, 2008 Revised manuscript due: January 28, 2008 Symposium: April 14-16, 2008 |
|
FURTHER INFORMATION |
|
Please
contact the Chairs: |
|
SPONSORS |
|
|
Monday, April 14
07:00 - 08:00 Speaker's breakfast
08:30 - 10:00 SpringSim Opening and Keynote Talk
Simulation and Cognitive Engineering:
The Use of Synthetic Environments to Design and Evaluate
Human-In-The-Loop Systems
The Visualization and Simulation (VSIM) Centre at Carleton University supports multidisciplinary research that integrates research in the human sciences with engineering. In the VSIM Advanced Cognitive Engineering (ACE) Lab, synthetic environments of different fidelity are engineered to support the design and evaluation of human-in-the-loop systems. The synthetic environments at the ACE Lab include HLA-enabled helicopter (CF-146 Griffon), general aviation (Cessna 172 type) and Unmanned Air Vehicle (UAV) simulators as well as an automobile simulator. In this talk he will outline a Cognitive Systems framework that we use to guide our research in the ACE lab and I will draw upon specific examples from our aerospace and automotive research to illustrate how simulation plays a critical role in advancing human-in-the-loop systems.
10:00 - 10:30 Coffee break
10:30 - 12:00 HPCS 2008 Keynote Presentation
The Need for Diversity in HPC
For both hardware and software in high-performance computing (HPC),
there are strong trends towards homogeneity. Although there are
strong economic, technological, and social reasons behind consolidation
in HPC, the lack of a diverse toolbox has some potential, long-term
risks. With only a single kind of hammer, will everything look
like a nail? If the tools to tackle the hard problems are no longer
available, will only the easy problems be solved?
We examine some of the trends in HPC hardware and software, the
tangible benefits that have resulted from consolidation, but also
the risks that come from a monoculture.
12:00 - 1:30 Lunch (on your own)
1:30 - 3:00 Paper Session 1
Towards Validation of DEVS-FIRE Wildfire Simulation Model
Solution of the Implicit Formulation of High Order Diffusion
for the Canadian Atmospheric GEM model
Optimizing Large Scale Chemical Transport Models for Multicore
Platforms
3:00 - 3:30 Afternoon Break
3:30 - 5:00 Paper Session 2
Porting a 3D image registration application to
multi-core environment
*HPCS 2008 Best Paper Award*
Constructing a Performance Database for Large-Scale Quantum
Chemistry Packages
Evening: SpringSim 2008 Reception
Tuesday, April 15
07:00 - 08:00 Speaker's breakfast
8:30 - 10:00 Tutorial Session
An Introduction to DEVS and Distributed DEVS
This tutorial presents fundamental DEVS concepts and applications of
DEVS-based modeling and simulation in industry and research. We will
discuss how DEVS is well suited to modern software infrastructures and
how it compares to non-DEVS modeling and simulation methodologies. In
particular, this tutorial will focus on distributed DEVS tools for
solving large-scale modeling and simulation problems.
10:00 - 10:30 Morning break
10:30 - 12:00 Paper Session 3
RMobiGen: A Trace Generation, Visualization,
and Performance Analysis Tool for Random Mobility Models
Performance Analysis of Web-based Distributed Simulation in DCD++:
A Case Study across the Atlantic Ocean
A Formal Framework for Stochastic DEVS Modeling and Simulation
12:00 - 1:30 Lunch/HPCS 2009 Planning Meeting
1:30 - 3:00 Paper Session 4
DEVS Session 2
An Approach for Simulation Based Structure Optimisation of
Discrete Event Systems
Parallel simulation of DEVS and Cell-DEVS models on Windows-based
PC cluster systems
Experiences with the DEVStone Benchmark
3:00 - 3:30 Afternoon Break
3:30 - 5:00 Panel session
Large Scale Scientific Computing in Canada
Wednesday, April 16
07:00 - 08:00 Speaker's breakfast
8:30 - 10:00 Tutorial Session
The Deterministic Global Optimization Algorithm DIRECT
The deterministic direct search algorithm DIRECT of D. Jones has
proven surprisingly practical for engineering design, especially when
coupled with a local pattern search algorithm such as MADS by Audet and
Dennis. Massively parallel implementations of these algorithms,
necessary for realistic engineering and scientific problems, are
highly nontrivial. This tutorial will cover the serial DIRECT and MADS
algorithms, their parallel implementations, place them in the context
of global optimization, and describe recent applications in
multidisciplinary design optimization and systems biology.
10:00 - 10:30 Morning break
10:30 - 12:00 Paper Session 5
High Performance Computing and Simulation Algorithms
A Shared Memory Parallel Algorithm for Data Reduction Using the
Singular Value Decomposition
Improving the Performance of Read-only Transactions
through Asyncronous Speculation
Fault-Tolerance and Reconfiguration of Circulant Graphs
and Hypercubes
12:00: HPCS 2008 concludes
Chris M. Herdman, Carleton University
Paul Lu, University of Alberta
Geophysical Applications
Session Chair: Adrian Sandu, Virginia Tech
Feng Gu (Georgia State University),
Xiaolin Hu (Georgia State University) and
Lewis Ntaimo (Texas A&M University)
Abdessamad Qaddouri and Vivian Lee (Environment Canada)
John Linford and Adrian Sandu (Virginia Tech)
High Performance Computing and Simulation Applications
Session Chair: Will Thacker, Winthrop College
Karoly Sandor (Budapest Tech), Miklos Kozlovszky (Budapest Tech),
Viktor Kamaras (3DHistech Ltd.), Levente Ficsor (3DHistech Ltd.),
Viktor Sebestyen Varga (3DHistech Ltd.) and
Bela Molnar (Semmelweis University)
Converting Macromolecular Regulatory Models from Deterministic
to Stochastic Formulation
Pengyuan Wang, Ranjit Randhawa, Clifford A. Shaffer, Yang Cao and
William Baumann (Virginia Tech)
Meng-Shiou Wu (Ames Lab), Hirotoshi Mori (Iowa State University),
Jonathan Bentz (Cray, Inc.), Theresa Windus (Iowa State University),
Heather Netzloff (Iowa State University), Masha Sosonkina (Ames Lab)
and Mark S. Gordon (Iowa State University)
Ming Zhang, University of Ottawa
DEVS Session 1
Session Chair: Trevor Pearce, Carleton University
Alex Aravind and Xiang Cui (University of Northern British Columbia)
Gabriel Wainer (Carleton University),
Qi Liu (Carleton University),
Julien Chazal (Université Blaise Pascal),
Loïc Quinet (Université Blaise Pascal)
and Mamadou K. Traoré (Université Blaise Pascal)
Rodrigo Castro (Universidad de Rosario), Ernesto Kofman (Universidad de Rosario)
and Gabriel Wainer (Carleton University)
Session Chair: Ming Zhang, University of Ottawa
Olaf Hagendorf and Thorsten Pawletta (University Wismar)
Bo Feng, Qi Liu and Gabriel Wainer (Carleton University)
Gabriel Wainer (Carleton University)
Moderator: Adrian Sandu (Virginia Tech)
Participants: Eric Aubanel (University of New Brunswick),
Waqar Haque (University of Northern British Columbia),
Vivian Lee (Environment Canada),
Gabriel A. Wainer (Carleton University),
and Ming Zhang (University of Ottawa)
Layne T. Watson, Virginia Polytechnic Institute and State University
Session Chair: Alex Aravind, University of Northern British Columbia
Rhonda Phillips, Layne Watson and Randolph Wynne (Virginia Tech)
Ragunathan Thirumalaisamy and Polepalli Krishna Reddy (IIIT, Hyderabad)
A. Farrag, S. Lou and Y. Qi (Dalhousie)