Colloques > Program

Linux Clusters Institute

4, 5 & 6 July 2012

HPC@LR organizes in Montpellier in partnership with IBM, "Mount Sinai Medical School", and the "Joint Institute of Computational Sciences" of the "University of Tennessee" and "Oak Ridge National Laboratory" (USA) a Summer School dedicated to High Performance Computing and Linux Clusters.

Worldwide specialists in Scientific Computing, Application development, Code porting and Tuning on various architectures, as well as Big Data and Linux Clusters will share their expertise with us.

See the Program for more details.

Computational science enables scientists to discover and explain the complex world around us. Through computational and data intensive simulations, the inner workings of the chemical and physical processes that control every scientific everything from the climate to the movement of the planets can be uncovered. To harness the power of high performance computing equipment and increase scientific discovery, we offer this workshop to bring participants up to date on the state-of-the-art in the field. We'll cover the basic motivations behind High Performance Computing (HPC), examine accelerator computing in depth, look at cloud computing and big data, and end the workshop with what HPC has done for the environment. Tutorials will allow participants to have hands on experience with HPC equipment, including accelerators.

We will begin the first day with an overview of HPC including parallel architectures and programming models. Then we will discuss the challenges and opportunities including the path of technology over the coming decade to computers capable of over 10^18 operations per second (exascale). In addition, we will also present success stories and application examples and provide an overview of the marketplace, including our expectations of future trends. Finally, we will examine application performance analysis and tools. The goal of the second day will be to introduce participants to GPUs and accelerator-based computing. We'll discuss programming models and the state of the market. On the third day, we will provide strategies for handling massive data sets and cover cloud computing. The workshop will end with a special session on how HPC applications have helped the planet.

For the first two days, we will have a tutorial using the equipment in the Montpellier Research Center, consisting of a xxx node Linux cluster, GPUs.... The tutorials will assist both beginners in running MPI and OpenMP codes, as well as advanced users who can bring their own application codes that we will assist in instrumenting with performance tools. We will also conduct a GPU tutorial.







Advised by: