Keynote and Invited Speakers 
Dr. Jack Dongarra | Dr.William Kramer | Dr. Peter Arzberger
Dr. Zhiwei Xu | Dr. Wen-Yih Sun | Dr. Mark Seager
Dr. Jack Dongarra
■ Abstract
In this talk we examine how high performance computing has changed over the last 10-year and look toward the future in terms of trends. These changes have had and will continue to have a major impact on our software. A new generation of software libraries and algorithms are needed for the effective and reliable use of (wide area) dynamic, distributed and parallel environments. Some of the software and algorithm challenges have already been encountered, such as management of communication and memory hierarchies through a combination of compile--time and run--time techniques, but the increased scale of computation, depth of memory hierarchies, range of latencies, and increased run--time environment variability will make these problems much harder. We will focus on the redesign of software to fit multicore architectures.
■ Short-Bio
Jack Dongarra received a Bachelor of Science in Mathematics from Chicago State University in 1972 and a Master of Science in Computer Science from the Illinois Institute of Technology in 1973. He received his Ph.D. in Applied Mathematics from the University of New Mexico in 1980. He worked at the Argonne National Laboratory until 1989, becoming a senior scientist. He now holds an appointment as University Distinguished Professor of Computer Science in the Computer Science Department at the University of Tennessee and holds the title of Distinguished Research Staff in the Computer Science and Mathematics Division at Oak Ridge National Laboratory (ORNL), Turning Fellow at Manchester University, and an Adjunct Professor in the Computer Science Department at Rice University. He is the director of the Innovative Computing Laboratory at the University of Tennessee. He is also the director of the Center for Information Technology Research at the University of Tennessee which coordinates and facilitates IT research efforts at the University.
He specializes in numerical algorithms in linear algebra, parallel computing, the use of advanced-computer architectures, programming methodology, and tools for parallel computers. His research includes the development, testing and documentation of high quality mathematical software. He has contributed to the design and implementation of the following open source software packages and systems: EISPACK, LINPACK, the BLAS, LAPACK, ScaLAPACK, Netlib, PVM, MPI, NetSolve, Top500, ATLAS, and PAPI. He has published approximately 200 articles, papers, reports and technical memoranda and he is coauthor of several books. He was awarded the IEEE Sid Fernbach Award in 2004 for his contributions in the application of high performance computers using innovative approaches and in 2008 he was the recipient of the first IEEE Medal of Excellence in Scalable Computing. He is a Fellow of the AAAS, ACM, and the IEEE and a member of the National Academy of Engineering.
---------------------------------------------------------------------------------------------------------------------top↑
Dr. William Kramer
■ Abstract
The next generation of Petascale systems presents incredible challenges in order to be able to effectively support science and engineering. This talk will focus on lessons learned fielding Terascale systems and discuss how these lessons can be applied to the Petascale era, using experiences from NERSC, NCSA and other systems with the example Petascale target platform of Blue Waters.
■ Bio Sketch
Dr. Kramer joined NCSA in 2008 as the Deputy Director of the Blue Waters Project. Blue Waters is under development and will be the largest unclassified system in the US in 2011. Supported by the US National Science Foundation and the University of Illinois, Blue Waters will provide a sustained Petaflops/sperformance for a wide range of applications of importance. As Deputy Director, Bill is responsible for all aspects of the Blue Waters project from the system itself to a range of collaborations that provide value added components to the system. Prior to this role, Bill was General Manager of the National Energy Research Scientific Computing Center (NERSC) at Lawrence Berkeley National Laboratory and Head of High Performance Computing for LBNL. Before that he was a Branch Chief for Computational Services at NASA's Numerical Aerodynamic Simulation (NAS) Facility. Bill is known for his experience in fielding early release, very large HPC systems that provide very high quality services to a very broad range of scientific domains. Blue Waters will is the 20th such system he is fielding. He has advanced degrees from Purdue University, University of Delaware and UC Berkeley.
---------------------------------------------------------------------------------------------------------------------top↑
Dr. Peter Arzberger
■ Abstract
One perspective of the High Performance Computing (HPC) community’s goal is to advance knowledge for the benefit of society. Starting from this perspective, we explore several examples from biomedical and environmental research that require different aspects of the HPC’s community’s technologies: computing, data, networking, and collaboration environments. We use these examples to provide model for the conduct of future research and education. Furthermore we explore the challenges of developing and providing infrastructure to conduct future research that will involve much more synthesis.
We will draw examples from activities from around the Pacific Rim, such as Pacific Rim Application and Grid Middleware Assembly (PRAGMA), that involves more than 30 institutions to build sustained collaborations and advance the use of grid technologies via applications; the National Biomedical Computation Resource (NBCR), an infrastructure resource for the biomedical community that focuses on targeted translational and multiscale challenges; and the Global Lake Ecological Observatory Network (GLEON), a grass-roots network of over 100 researchers who have a common goal of building a scalable, persistent network of lake ecology observatories. In addition we will introduce three programs aimed at educating the next generation of researchers in international research apprenticeships: Pacific Rim Experineces for Undergraduates (PRIME), Pacific Rim International UniverSity (PRIUS), and Monash Undergraduate Research Program Abroad (MURPA).
We hope that the material we present will challenge the HPC community in its future directions, to the benefit of society.
■ Bio Sketch
Peter Arzberger is Chair of the Pacific Rim Application and Grid Middleware Assembly (PRAGMA; www.pragma-grid.net), an open, institution-based organization of more than 30 institutions. PRAGMA, founded in 2002, has a mission to build sustained collaborations among researchers around the Pacific Rim by building applications on top of emerging Grid hardware and software. Build on the foundation of PRAGMA is PRIME, the Pacific Rim Undergraduate Experiences (prime.ucsd.edu) program, which provides international research and cultural internship experiences to undergraduate students. PRIME, founded in 2004, has admitted 36 students and sent students to four PRAGMA sites. Arzberger is a founding member of the Steering Committee another international activity, GLEON (http://www.gleon.org), the Global Lake Ecological Observatory Network. GLEON is a grassroots network of people, institutions, programs, and data linked by cyberinfrastructure and united by the mission to understand and predict the response of lake ecosystems to natural processes and human activities at regional, continental, and global scales.
In addition, Arzberger is Director of the National Biomedical Computation Resources (http://nbcr.net), an NIH National Center for Research Resource award. NBCR’s mission is to conduct, catalyze and enable biomedical research by harnessing forefront computational and information technologies, focused on targeted translational and multiscale challenges. He is also Chair of the National Advisory Board to the U.S. Long Term Ecological Research (LTER) network. Arzberger is the former Executive Director of the National Partnership for Advanced Computational Infrastructure (NPACI) and a former Program Officer at the United States National Science Foundation in Computational Biology.
--------------------------------------------------------------------------------------------------------------------top↑
Dr. Zhiwei Xu
■ Abstract
This talk presents the growth and challenges of high performance computing in China. In particular, we trace the development of Dawning superservers, including the newly announced Dawning 5000 systems based on x86 or Godson microprocessor chips. Issues of cost, power, heat, efficiency, and ease of use and maintenance are discussed, as well as their impacts on different users’ adoptions of the high performance computing technology. Results are illustrated on how the Dawning systems have been designed to alleviate these problems. Research activities are reviewed on workload classification and characterization, the scalable architecture of Godson 3 microprocessor chip design, the HPP (Hyper Parallel Processing) architecture of Dawning 5000, and the system software stack. Based on historical trends, we argue that personal high performance computing systems are needed to broaden the user community. Progress is reported on low-cost high performance computing for the masses. An example is given for Dawning HPW, a personal high performance computing system series towards the performance/cost target of 1 Tflop/s per $50 thousand.
■ Bio Sketch
Zhiwei XU received his PhD degree from University of Southern California in 1987. He is a professor and CTO of the Institute of Computing Technology (ICT) of Chinese Academy of Sciences. His research areas include net computing, operating systems, and high-performance computer architecture. He serves on the editorial boards of several international journals, such as IEEE Transactions on Computers and Journal of Parallel and Distributed Computing.
---------------------------------------------------------------------------------------------------------------------top↑
Dr. Wen-Yih Sun
■ Abstract
British mathematician Lewis Fry Richardson first proposed numerical weather prediction in 1922. He attempted to perform a numerical forecast but failed. The first successful numerical simulation was performed in 1950 by a team composed of Charney, Thompson, Gates, Fjörtoft and Neumann, using the ENIAC digital computer. They used a simple barotropic vorticity equation. Later models used more complete equations for atmospheric dynamics and thermodynamics. Currently, most research and operational centers use super computers to simulate and/or predict the weather and climate, including typhoon, flood, drought, etc. One of the famous examples is Earth Simulator in Yokohama, Kanagawa, Japan which is capable of 35.86 trillion (35,860,000,000,000) floating-point calculations per second, or 35.86 TFLOPS.
A numerical weather predication model is based upon the Navier-Stokes equations, which consists of the prognostic equations for temperature, wind, moisture, cloud, rain, snow, soil property etc. These equations are nonlinear and are impossible to solve exactly. Therefore, numerical methods obtain approximate solutions. Different models use different initial/boundary conditions, resolutions, numerical methods, or different physics, which may result in different forecasting. Usually, a model with better resolutions, more comprehensive physics and sophisticated numerical methods produces better results. Furthermore, closure problems exist in sub-grid scale turbulence. Hence, more powerful computers are always required in numerical weather prediction. In addition to sub-grid scale turbulence, uncertainty also exists in the initial and boundary conditions, as well as in physics parameterizations of the model. The errors also exist in the numerical schemes. Therefore, ensemble forecasting has been performed for the past few years, which also requires super-computers (to run the different models with different conditions). Taiwan has a very complicated terrain, (which requires the fine resolution model). It is located in the transition between tropical and mid-latitude regions, as well as the most favorable zone for typhoon passages, (which needs a large domain with comprehensive dynamics and physics). Hence, it is a great challenge to develop a comprehensive forecasting model as well as to write the parallel code with high efficiency to predict the weather/climate and flood/drought in Taiwan and surrounding areas, which requires a close collaboration between the NCHC and TTFRI. The collaboration is also required in preparing and handling the huge initial data and analyzing/displaying the numerical results.
■ Short-Bio
● Current Position:
Professor of Earth and Atmospheric Sciences, Purdue University (since 1988)
Director, the (Preparatory) National Center for Typhoon and Flooding Research, Taiwan
University Chair Professor, Department of Atmospheric Sciences,
National Central University, Chung-Li, Taiwan
University Chair Professor, Institute of Environmental Engineering, National Chiao Tung University, Hsin-chu, Taiwan.
● Education:
Ph.D. (1975): The University of Chicago, Geophysical Sciences
MS (1972): The University of Chicago, Geophysical Sciences
BS (1968): National Taiwan University, Atmospheric Sciences
● Professional Experience:
May - August 2004:
Visiting Professor, Center for Climate System Research, University of Tokyo, Japan
January - July 2002:
Visiting Professor, National Taiwan University
May - June 1996:
Visiting scientist, Meteorological Institute Stockholm University, Stockholm, Sweden
1995 -1996:
Visiting Professor, National Taiwan University & National Normal University, Taiwan
1988 - present:
Professor of Earth and Atmospheric Sciences, Purdue University
1984 -1988:
Associate Professor of Earth and Atmospheric Sciences, Purdue University
1979 -1984:
Assistant Professor of Earth and Atmospheric Sciences, Purdue University
1978 -1979:
Visiting Scientist, Geofluid Dynamics Program, Princeton University
1975 -1978:
Research Assistant Professor, Laboratory for Atmospheric Research, University of Illinois
● Specialty
Geofluid dynamics, weather and regional climate modeling
---------------------------------------------------------------------------------------------------------------------top↑
Dr. Mark Seager
■ Abstract
Building, integrating and using petascale systems have many challenges including system power and cooling, system stability, scalablity, simulation environment and the development of petascale applications. In this talk, we discuss these challenges and provide some approaches to addressing these challenges. In addition, we discuss some recent scientific results from petascale systems that make the whole effort worthwhile.
■ Bio Sketch
Dr. Seager received his B.S. Degree in Mathematics and Astrophysics at the University of New Mexico at Albuquerque in 1979 and received his PhD in Numerical Analysis from the University of Texas at Austin in 1984. Dr. Seager started working at Lawrence Livermore National Laboratory in 1983 and has been working in the field of parallel processing ever since. He manages the Platforms Program for the Advanced Simulation and Computing (ASC) Program at LLNL and has managed multiple vendor partnerships to successfully procure, deploy and integrate architectures such as ASCI Blue Pacific (3.9 TF/s in 1998), ASCI White (12.3 TF/s in 2000) and Purple (100TF/s in 2005) and BlueGene/L (360 TF/s in 2005). In addition, Dr. Seager developed the LLNL Linux strategy and helped deploy multiple generations of leading edge clusters (MCR at 11.3 TF/s in 2002 and Thunder at 23 TF/s in 2004) In 2006 when faced with the challenge of deploying multiple Linux clusters of various sizes per year while at the same time reducing total cost of ownership by 50%, Dr. Seager developed the Scalable Unit concept. With the Peloton and TLCC07 procurement activities, this strategy delivered over 20 clusters in excess of 600 TF/s aggregate from 35 scalable units to three USA National Laboratories. Dr. Seager recently led the ASC Sequoia procurement for a 20 PF/s system to be delivered in 2011. Dr. Seager is now focused on the challenges of peta-scale systems, simulation environments and applications development strategies.
---------------------------------------------------------------------------------------------------------------------top↑
