Vincent Keller
EPFL ENT-R SCITAS
ME B2 444 (Bâtiment ME)
Station 9
CH-1015 Lausanne
Mission
- analyze the needs of EPFL and UNIL (University of Lausanne) applications on large scale machines
- optimize and/or parallelize scientific applications (500+ speedups have been measured)
- advise scientists with best practices in HPC programming
- find new researchers who wants to obtain a jump of scale with their simulations
- connect people from different fields with different skills for common projects (ex: I managed a CS student for a Master project in computational mechanics with experts in this domain)
- manage the time sharing of the 15+ production projects on the centers IBM BG/Q supercomputer
- share HPC application analysis and energy efficiency expertise for EPFL central resources acquisitions
- in-service formation with bleeding edge and up-to-date HPC technologies
Biography
Vincent Keller is a computer scientist born in 1975 in Lausanne, Switzerland. He received his Master degree in Computer Science from the University of Geneva (Switzerland) in 2004. From 2004 to 2005, he holds a full-time researcher position at the University Hospital of Geneva (HUG). He was involved at HUG on simulating blood flows in cerebral aneurysms using real geometries constructed from 3D X-rays tomography slices. The numerical method used was Lattice-Boltzman Method (LBM). He received his PhD degree in 2008 from the École Polytechnique Fédérale de Lausanne (EPFL) in the HPCN and HPC Grids fields. His supervisor was Dr. Ralf Gruber. He developed a low-level application-oriented monitoring system (VAMOS) and the Resource Broker of the ïanos (Intelligent ApplicatioN-Oriented System) framework prototype. In 2009, he holds a full-time researcher position at Fraunhofer SCAI (University of Bonn) in Germany. He was the Scientific Coordinator of the IANOS project. In 2010, he was application analyst in the IT service at University of Zürich. Since September 2010, he is application analyst in the Center for Advanced Modeling Science (CADMOS) for the EPFL&University of Lausanne projects. Dr. Vincent Keller is the author or coauthor of more than 30 peer-reviewed papers and one book (HPC@GreenIT, Springer Verlag, Heidelberg, Germany, October 2009). His research interests are in HPC applications analysis, Grid and cluster computing and energy efficiency of large computing ecosystems.
Websites
Visit the Center for Advanced Modeling Science website
Fields of expertise
High Performance Computing, Computational Science and Engineering Parallel computing (over distributed memory, shared memory or hybrid environments)
Optimization, Parallelization, Parallel I/O, Scientific Libraries
Scientific Applications and Performance Analysis, Algorithms
Energy Efficiency in Large Computing Ecosystems
Archival processing
Wine tasting
Publications
MAIN PUBLICATIONS
Vincent Keller, Kevin Cristiano, Ralf Gruber, Pierre Kuonen, Sergio Maffioletti, Nello Nellari, Marie-Christine Sawley, Trach-Minh Tran, Philipp Wieder, Wolfgang Ziegler Proc. of CoreGRID Integration Workshop. CoreGRID 2005, Pisa, Italy |
Integration of ISS into the VIOLA Meta-scheduling Environment |
R. Gruber, V. Keller, P. Kuonen, M.-Ch. Sawley, B. Schaeli, A. Tolou, M. Torruella, and T.-M. Tran Proc. of 6th Int. Conf. PPAM 2005, Poznan, Poland, Lecture Notes in Computer Science 3911 (Springer, 2006) 751-757 |
Towards an Intelligent Grid Scheduling System |
Vincent Keller, Michela Thiémard Flash Informatique, Spécial AlterIT, Ete 2005 |
Des petits hommes verts au Numéro 1 du Top500 |
R. Gruber, V. Keller, E. Leriche Proc. of Cluster2006 Conf, 2006, Barcelona, Spain |
Can a Helmholtz solver run on a cluster |
Ralf Gruber , Vincent Keller , Michela Spada Proc of UniCORE Summit 2006, Europar, Dresden, Germany |
Integration of Grid Cost Model into ISS/VIOLA |
Infoscience
Teaching & PhD
Teaching
- Mathematics
PhD Programs
PhD Students
Courses
Parallel and high-performance computing
This course provides insight into a broad variety of High Performance Computing (HPC) concepts and the majority of modern HPC architectures. Moreover, the student will learn to have a feeling about what architectures are suited for several types of algori...
Parallel programming
Learn the concepts, tools and API's that are needed to debug, test, optimize and parallelize a scientific application on a cluster from an existing code or from scratch. Both OpenMP (shared memory) and MPI (distributed memory) paradigms are presented and...