Research

My current research revolves around HPC. The holy Grail is now the exascale (10¹⁸ Flop/s). To achieve this goal, the newly built, and to-be built, supercomputers expose more and more heterogeneity in the way there are built. My main focus is the design of scheduling algorithms taking into account these architectural evolutions.

I am interested in designing low complexity scheduling algorithms with guaranteed performances. I am targeting the fine grain scale (i.e., single node machines) as well as coarse grain (i.e., whole machine scale, roughly 10⁶ nodes).

My work covers both the theoretical aspects (design of algorithms, complexity study, performance guarantees) and the practical aspects (implementation, performance evaluation through simulation and experimentation).

Publications

Please find below the comprehensive list of my publications. If interested, you may subscribe to a RSS feed listing my publications. A comprehensive list of my publications is also available as a .bib file.

Journals

Raphaël Bleuse, Sascha Hunold, Safia Kedad-Sidhoum, Florence Monna, Grégory Mounié, and Denis Trystram. Scheduling Independent Moldable Tasks on Multi-Cores with GPUs. IEEE Transactions on Parallel and Distributed Systems, 28(9):2689–2702, 2017. doi:10.1109/TPDS.2017.2675891, hal:hal-01516752.

Raphaël Bleuse, Safia Kedad-Sidhoum, Florence Monna, Grégory Mounié, and Denis Trystram. Scheduling independent tasks on multi-cores with GPU accelerators. Concurrency and Computation: Practice and Experience, 27(6):1625–1638, 2015. doi:10.1002/cpe.3359, hal:hal-01081625.

Conferences/Workshops

Rosa Pagano, Sophie Cerf, Bogdan Robu, Quentin Guilloteau, Raphaël Bleuse, and Éric Rutten. Making Control in High Performance Computing for Overload Avoidance Adaptive in Time and Job Size In CCTA. IEEE, 2024. hal:hal-04669743.

Ismail Hawila, Sophie Cerf, Raphaël Bleuse, Swann Perarnau, and Éric Rutten. Adaptive Power Control for Sober High-Performance Computing. In CCTA, 403–410. IEEE, 2022. doi:10.1109/CCTA49430.2022.9966115, hal:hal-03765849.

Sophie Cerf, Raphaël Bleuse, Valentin Reis, Swann Perarnau, and Éric Rutten. Sustaining Performance While Reducing Energy Consumption: A Control Theory Approach. In Euro-Par, volume 12820 of Lecture Notes in Computer Science, 334–349. Springer, 2021. doi:10.1007/978-3-030-85665-6_21, arxiv:2107.02426, hal:hal-03259316.

Maya Olszewski, Jeff Meder, Emmanuel Kieffer, Raphaël Bleuse, Martin Rosalie, Grégoire Danoy, and Pascal Bouvry. Visualizing the Template of a Chaotic Attractor. In Graph Drawing, volume 11282 of Lecture Notes in Computer Science, 106–119. Springer, 2018. doi:10.1007/978-3-030-04414-5_8, arxiv:1807.11853, hal:hal-02099518.

Raphaël Bleuse, Konstantinos Dogeas, Giorgio Lucarelli, Grégory Mounié, and Denis Trystram. Interference-Aware Scheduling using Geometric Constraints. In Euro-Par, volume 11014 of Lecture Notes in Computer Science, 205–217. Springer, 2018. doi:10.1007/978-3-319-96983-1_15, hal:hal-01884542.

Raphaël Bleuse, Giorgio Lucarelli, and Denis Trystram. A Methodology for Handling Data Movements by Anticipation: Position Paper. In Euro-Par Workshops, volume 11339 of Lecture Notes in Computer Science, 134–145. Springer, 2018. doi:10.1007/978-3-030-10549-5_11, hal:hal-02066200.

Raphaël Bleuse, Thierry Gautier, João Vicente Ferreira Lima, Grégory Mounié, and Denis Trystram. Scheduling Data Flow Program in XKaapi: A New Affinity Based Algorithm for Heterogeneous Architectures. In Euro-Par, volume 8632 of Lecture Notes in Computer Science, 560–571. Springer, 2014. doi:10.1007/978-3-319-09873-9_47, arxiv:1402.6601, hal:hal-01081629.

Theses

Raphaël Bleuse. Apprehending heterogeneity at (very) large scale. PhD thesis, LIG – Univ. Grenoble Alpes, 2017. hal:tel-01722991.

Raphaël Bleuse. Utilisation efficace des accélérateurs GPU – Ordonnancement sur machines hybrides. Master's thesis, Ensimag – Grenoble INP, 2013. hal:hal-00858233.