Presenter

Marc Pérache is a research engineer at “Commissariat à l’Energie Atomique et aux Energies Alternatives”. He has been working in the HPC field since 2003, starting with a Ph.D. on runtime systems for cluster of NUMA nodes (received in 2006). During his Ph.D., he started the MPC (MultiProcessor Computing) framework. He joined CEA in 2006 as a research engineer to extend the MPC framework and provide building blocks for high performance multithreaded applications. He received the “Habilitation à Diriger des Recherches” (French degree which accredits to supervise researchers) from Versailles Saint Quentin-en-Yvelines University in 2015. Since 2018, he leads an R&D group dealing with runtime systems, tools for HPC systems (MPI, OpenMP, memory allocation, high speed networks, scalable profiling tools …) and framework for scientific computing codes.