Research topics

Since 2004, my research activities within the Dolphin project-team (Inria Lille – CRIStAL) have primarily focused on cooperative parallel optimization for large-scale platforms, in particular computational grids. Several issues have been addressed, including checkpointing, optimized encoding of work units to cope with wide-area communications, and dynamic adaptive load balancing. My research activities have then evolved towards parallel optimization on heterogeneous environments, including multi-core processors and accelerators such as GPUs.

The main challenges addressed include thread divergence, memory optimization, CPU–GPU data transfer optimization, and the definition of new data structures to efficiently store and manage large amounts of generated subproblems. Within the context of the BONUS team, since 2017, my interests focus on big optimization, i.e., high-dimensional and expensive optimization, using decomposition, surrogate-based machine learning, and extreme-scale computing. Large clusters involving multi-core processors and GPU accelerators are mainly considered. More recently, I have also integrated quantum optimization into my research activities, both through the use of parallel optimization for NISQ systems, such as quantum circuit transpilation, and through the exploration of hybrid HPC–quantum approaches for solving challenging optimization problems. From an application perspective, I am interested in aerospace design engineering, as well as Tuberculosis and COVID-19 transmission control in epidemiology.