SELECT * FROM pwn_ihpcf_person_info WHERE ihpcfusername = 'Petiton' Serge PetitonProfessor, University of Lille; Senior Researcher, Maison de la Simulation, CNRS

Serge Petiton

Professor, University of Lille; Senior Researcher, Maison de la Simulation, CNRS

Prof. Serge G. Petiton received the M.S. in Applied Mathematics, the Ph.D. degree in computer science, and the “Habilitation à diriger des recherches”, from Pierre and Marie Curie University, Univ. PARIS 6. He was post-doc student, registered at the graduate school, and junior researcher scientist at the YALE University, 1989-1990. He has been researcher at the “Site Experimental en Hyperparallelisme” (supported by CNRS, CEA, and French DoD) from 1991 to 1994. He also was affiliate research scientist at YALE and visiting research fellow in several US laboratories, especially in NASA-ICASE and the AHPCRC during the period 1991-1994. Since then, Serge G. Petiton is Professor at the University of Lille, Sciences and Technologies. Serge G. Petiton was, and is, P.I. of several international projects with Japan, Venezuela and Germany. Since 2012, Serge G. Petiton has a half time CNRS senior position at the “Maison de la Simulation” in Saclay. Serge G. Petiton has been scientific director of more than 22 Ph.D.s and has authored more than 100 articles on international journals and conferences. His main current research interests are in “Parallel and Distributed Computing”, “Post-Petascale Auto/smart-tuned Dense and Sparse Linear Algebra”, and “Language and Programming Paradigm for Extreme Modern Scientific Computing”; targeting especially geoscience and big data applications. Serge G. Petiton is a member of SIAM, ACM, IEEE, the YALE club of France and the Ivy Plus European leaders club.

Talk Title:

Graph of components and containers programming for extreme computing

Talk Abstract:

Future supercomputers are expected to have highly hierarchical architectures with nodes composed by lot-of-core processors and accelerators. The different programming levels will generate new difficult algorithm issues. New intelligent applications would mix computational and data sciences. Moreover, methods have to be redesigned and new ones introduced or rehabilitated, in particular in terms of communication optimizations and data distribution. Then, new languages and frameworks should be defined and evaluated with respect to modern state-of-the-art of scientific methods. In this talk, we first present a solution based on graph of components programming experimented on several machines for dense and sparse linear algebra. Hence, we focus on YML with its high level language allowing to automate and delegate the managements of dependencies between loosely coupled clusters of processors to a specialized tool controlling the execution of the application using parallel tasks. We present some unite and conquer approaches for scientific computation well-adapted for such programming paradigm and we discuss new researches mixing components and containers associated with data science computation. We conclude proposing the integration of several obtained results toward intelligent computing on the road to exascale and beyond.

Update my resume!