The International Conference on High Performance Computing & Simulation (HPCS 2012) or

July 2 – July 6, 2012

Madrid, Spain

In Cooperation with ACM, IEEE, and IFIP



Energy Efficient Systems in Next Generation High Performance Data and Compute Centers


Jaafar Gaber Université de Technologie de Belfort-Montbéliard (UTBM), France

Vicente Martin Director of CESVIMA, Polytechnic University of Madrid, Spain

Miguel A. Ordoñez IBM Global Services España, Madrid, Spain

Johnatan Pecero University of Luxembourg, Luxembourg


Jesus Carretero and Pascal Bouvry

Universidad Carlos III de Madrid, Madrid, Spain; University of Luxembourg, Luxembourg


As large scale distributed systems gather and share more and more computing nodes and storage resources, their energy consumption is exponentially increasing. Next generation computing and data centers might require tens of MWatts to be feasible. Thus designing more efficient systems is a major challenge for computer engineers. This challenge is two-folds: saving money and being ecological by using renewable energy.

The goal of this panel is to discuss current trends in energy use and energy costs of data centers and servers, metrics to benchmark computing and data centers energy consumption, and opportunities for reducing those costs through improved hardware and software.


Performance Modeling and Simulation for System Design, Evaluation, and Optimization in the Era of Exascale Sciences: Challenges, Needs and Requirements


David R.C. Hill ISIMA / LIMOS – Blaise Pascal University, Clermont-Ferrand, France

Jesus Labarta Barcelona Supercomputing Center & Technical University of Catalonia (UPC), Spain

Joseph Schneible HPC laboratory, The George Washington University, USA

James C. Sexton IBM T. J. Watson Research Center, Yorktown Heights, NY, USA


Mads Nygård Norwegian University of Science and Technology, Trondheim, Norway


Computer simulation, or simulation-based engineering and science, is now widely accepted by many companies as a way of testing the product before manufacturing physical prototypes. It is considered as an integral part of the product development process allowing scientific and technological innovation and discovery in many areas such as the design and the development of next generation combustion systems and virtual prototyping and product design. Using simulation-based engineering design, companies are expected to deliver more business value to their end customers in the dynamic business changing environment by reducing product development cost and time while increasing the performance.

There are many similar gains in other fields, such as healthcare, biomedical and biosciences, climate and environmental changes, design and manufacturing of advanced materials, geology, chemistry and physics, and even financial systems.

However, in order to achieve these goals, systematic product development methods for simulation-based science and engineering designs are required. Among the issues to be tackled are the management of massive amount of parallel simulations of separate computer-aided engineering applications and simulations using complex mathematical algorithms that could take long computational time with less accuracy. Therefore, systematic product development methods for simulation driven design that address these issues and others are particularly required.

Furthermore, despite existing HPC techniques, tools, and platforms for scientific applications acceleration, companies are hesitating to introduce HPC platforms into their engineering simulation environments because of cost and complexity. This will become more evident as we march toward the Exascale era.

Recent directions in research and technological studies have shown that Exascale computing may provide solutions for some of these issues by leading to advances in several areas of science and technology while allowing more computationally intensive and data intensive problems to be handled successfully. However, several computational and technical challenges, from hardware (e.g., data access/storage hierarchy) to algorithms, to programming models, and applications, must be overcome to make Exascale computing a practical infrastructure for computer-aided engineering applications and simulations.

In this panel, we shall discuss the promise of Exascale computing to transform the role of modeling and simulation in science and engineering. Panelists will highlight research directions for developing Exascale computing infrastructures including models to support new emerging architectures, programming models to support exasclae computing, tools to simulate and evaluate these architectures, and programming languages (hybrid or new languages) that are appropriate for these new architectures. The intent is to make future HPC widely accessible to the designers, engineers, and scientists, in order for them to focus on how to design and simulate their systems instead of focusing on the computing systems issues.