Find Academic Supervisors

• Big Data management
• Cloud computing
• Distributed systems
• +3 more

Shadi Ibrahim is a Researcher at the Inria Centre at Rennes University , specializing in Big Data management since 2013. His work focuses on optimizing data flow processing in cloud environments, particularly through the KerStream project (2017–2022), which addressed limitations in platforms like Spark and Flink by developing methods for task lag detection and energy-efficient allocation. Key research areas: Big Data, Cloud computing, Erasure codes, HPC convergence Notable collaborations include industry partnerships with OVH and startup Hive (2021–2022) to reduce data duplication via erasure codes. He received the ACM Distinguished Member honor (2022) and the IEEE TCSC Excellence Prize (2020). His work also intersects with exascale computing initiatives via partnerships with Argonne National Laboratory and Lawrence Berkeley National Laboratory .

• Turbulent Combustion
• Numerical Simulation
• Heat Transfer
• +3 more

Ronan Vicquelin is a University Professor (1st Class) at CentraleSupélec, Paris-Saclay University, affiliated with the EM2C Laboratory (CNRS). He serves as Head of the Department of Aeronautics, Space and Transport and co-supervises the High Performance Computing Mésocentre. His academic appointments include previous roles as University Professor (2nd class) and Head of Aerospace programs. Education includes Habilitation (University of Rouen Normandy, 2018), PhD in Energetics (École Centrale Paris, 2010), M.Sc. in Mechanical Engineering & Aerospace (École Centrale Paris, 2006), and Engineering Diploma (École Centrale Paris, 2006). Research focuses on turbulent reacting flows with emphasis on: numerical simulation of combustion systems, LES/DNS methodologies, uncertainty quantification, hydrogen combustion dynamics, conjugate heat transfer, and radiative energy transfer. Current investigations explore flame stabilization mechanisms, multi-physics coupling, and high-performance computing applications for aerospace propulsion systems. Publications predominantly address combustion science, with recent works (2021-2025) emphasizing hydrogen flame dynamics, NOx emission control, advanced numerical methods for reactive flows, and experimental validation of turbulent combustion models. Thermal radiation effects and multi-phase flow interactions constitute emerging themes. Advises multiple PhD candidates with projects funded by ANR, EU programs (ACHIEVE, SOPRANO), and industry partnerships (Safran, Air Liquide). Research grants include PEPR OXY3C, ANR HyMaX, and ANR OXYTEC focusing on zero-emission combustion technologies. Leads experimental and computational research at EM2C Laboratory, coordinating teams working on turbulent combustion diagnostics, high-fidelity simulations, and development of the Mésocentre HPC infrastructure for large-scale CFD.

Marc SnirView profile

Professor

• Parallel Computing
• Computer Architecture
• High Performance Computing
• +3 more

Marc Snir is an Israeli-American Professor of Computer Science at the University of Illinois at Urbana-Champaign (UIUC). He holds a Doctor Honoris Causa from École Normale Supérieure de Lyon (2018). His career includes leadership roles such as Head of the Computer Science Department at UIUC (2001–2007) and Director of the Mathematics and Computer Science Division at Argonne National Laboratory (2011–2016). He is a Fellow of AAAS, ACM, and IEEE, and recipient of the IEEE Seymour Cray Award. Education: PhD in Mathematics from the Hebrew University of Jerusalem (1979). Research focuses on parallel algorithms, high-performance computing (HPC), distributed systems, and fault tolerance. He co-developed the MPI standard for parallel programming and contributed to IBM's Blue Gene supercomputer architecture. His Erdős number is 2, reflecting collaborative links to mathematician Paul Erdős. Awards include the IEEE Award for Excellence in Scalable Computing and leadership in advancing supercomputing through publications like The Future of Supercomputing (2004).

Samuel ThibaultView profile

Professor

• Parallel Computing
• High Performance Computing
• Task-Based Runtime Systems
• +3 more

Samuel Thibault is a Professor at University of Bordeaux, affiliated with Laboratoire Bordelais de Recherche en Informatique (LaBRI) and Inria Bordeaux -- Sud-Ouest. He is a member of the SATANAS team at LaBRI (theme: High Performance Runtime Systems for Parallel Architectures) and the STORM research team at Inria Bordeaux (previously RunTime). His work bridges academic research and practical implementation of high-performance computing systems. Thibault's research focuses on task-based runtime systems, particularly StarPU, for heterogeneous and parallel computing architectures. His work addresses critical challenges in scheduling algorithms, memory management under constraints, data locality optimization, and performance modeling for complex NUMA architectures. He has made significant contributions to the field of parallel computing through the development and analysis of runtime systems that efficiently manage tasks across diverse hardware resources including CPUs, GPUs, and other accelerators. His research has practical applications in scientific computing, deep learning inference, and large-scale simulations requiring extreme computing power. His recent publication trends show a strong emphasis on optimizing task-based runtime systems for heterogeneous architectures with particular attention to memory constraints and data locality. There's a clear progression toward applying these techniques to deep learning inference workloads, as evidenced by his StarONNX project. His work consistently addresses the challenge of balancing throughput and latency in complex computing environments, with increasing focus on recursive task graphs and dynamic adaptation strategies. Thibault is actively involved in European research initiatives including TEXTAROSSA (focusing on exascale technologies) and EXA2PRO (high development productivity on heterogeneous systems). His work has been published consistently in top-tier conferences and journals in parallel and distributed computing. As part of the STORM team at Inria Bordeaux, Thibault contributes to advancing the state of the art in runtime systems for high-performance computing. His team's work on StarPU has become a reference implementation in the field, enabling researchers and practitioners to develop applications that can efficiently utilize heterogeneous computing resources without needing to manage the complexity of different hardware architectures directly.

François TrahayView profile

Professor

• High Performance Computing
• Runtime Systems
• Performance Analysis
• +4 more

François Trahay is a Full Professor in the Computer Science department at Télécom SudParis (Institut Mines-Télécom) and a member of the Benagil Inria team. He leads research in high-performance systems, runtime systems, and performance analysis for HPC and distributed systems. He holds an HDR from Institut Polytechnique de Paris and a PhD from University of Bordeaux (2009). His work includes the EZTrace framework for performance analysis and contributions to storage systems optimization. Education: 2021: Habilitation à Diriger des Recherches (HDR), Institut Polytechnique de Paris 2010: PostDoc at Riken, University of Tokyo 2009: PhD in Computer Science, University of Bordeaux 2006: MS in Computer Science, University of Bordeaux Research focuses on runtime system design, HPC trace analysis, and storage efficiency. Recent projects include PALLAS trace format (IPDPS 2025) and GPU performance prediction (Euro-Par 2024). He advises 4 current PhD students and has supervised 2 former students now in industry. Key contributions include: Co-developer of EZTrace performance analysis framework Co-author of 60+ peer-reviewed papers in top venues (IPDPS, IEEE Cluster, ICPP) Technical leadership in Inria's Benagil team and Samovar Lab

Francois TrahayView profile

Associate Professor

• High Performance Computing
• Storage Systems
• Parallel and Distributed Systems
• +2 more

François Trahay is an Associate Professor at Telecom SudParis, affiliated with the SAMOVAR research laboratory. His research focuses on high-performance computing systems, storage optimization, and performance analysis tools for parallel and distributed environments. He completed his PhD at Université Bordeaux I (2009) and Habilitation (HDR) at Institut Polytechnique de Paris (2021). Research Interests: Trahay specializes in optimizing storage systems (SSDs, RAID configurations), developing performance analysis frameworks (e.g., EZTrace, NumaMMA), and enhancing energy efficiency in HPC. Key areas include I/O performance, parallel runtime systems, and adaptive computing for machine learning workloads. Publication Trends: His recent work (2018–2025) demonstrates strong focus on: (1) SSD/RAID management techniques for modern storage hardware, (2) HPC performance tools for tracing and analysis, and (3) optimization strategies for distributed deep learning systems. Laboratory Affiliation: Member of SAMOVAR Laboratory (UMR CNRS), conducting research in distributed systems, networks, and computational efficiency.

Guillaume HuardView profile

Associate Professor

• High Performance Computing
• Applications Traces
• Monitoring & Collection Aggregation & Visualization Runtime
• +5 more

Guillaume Huard is an Associate Professor at Grenoble Alpes University, affiliated with the LIG Laboratory and the POLARIS Research Team . His research focuses on high-performance computing (HPC), particularly in tracing, performance analysis, and optimization of parallel applications, compiler optimization, and architecture. PhD in 2002 from École Normale Supérieure de Lyon under Alain Darte His research interests span applications tracing, monitoring and visualization of runtime data, application deployment, scheduling, performance evaluation, resource utilization, and cache/memory footprint optimization. His work often addresses challenges in NUMA architectures and large-scale trace analysis. Guillaume's publications highlight trends in memory trace collection (e.g., Moca system), spatiotemporal data aggregation, NUMA-aware visualization tools (e.g., TABARNAC), and cache utilization control. Key subfields include performance debugging, compiler transformations, runtime environments, and multidimensional data aggregation techniques. Guillaume is actively involved in the POLARIS Bootcamp and DATAMOVE/POLARIS events , contributing to collaborative research and educational initiatives in HPC.

Anne BENOITView profile

Associate Professor

• High-Performance Computing (HPC)
• Multi-criteria scheduling algorithms
• Resource optimization
• +4 more

Anne BENOIT is an Associate Professor of Computer Science at École normale supérieure de Lyon (ENS de Lyon), affiliated with the Parallel Computation Laboratory (LIP) and the ROMA project. She holds leadership roles, including Chair of the IEEE Technical Community on Parallel Processing (2020–2024) and IUF Senior Membership (2023–2028). Her research focuses on multi-criteria scheduling algorithms for HPC, addressing performance, cost, energy, and reliability challenges. She earned a PhD in Systems and Communications from INPG Grenoble (2003) and an HDR (2009). Research Interests : Her work integrates scheduling optimization, fault tolerance, and energy efficiency in HPC systems. She designs resilient protocols to manage failures and uncertainties, particularly in heterogeneous environments. Key topics include checkpointing strategies, workflow scheduling on edge platforms, and balancing conflicting optimization criteria. Awards & Roles : Senior Member of the Institut Universitaire de France (2023) Guest Professor at Georgia Tech (2017–2018) Associate Editor-in-Chief of Elsevier JPDC and ParCo IEEE Senior Member Professional Contributions : Chair of major conferences like IPDPS and HiPC. Authored A Guide to Algorithm Design (CRC Press, 2013). Leads the Computer Science Department at ENS Lyon since 2022. Labs & Projects : Active in the ROMA project (Resource Optimization) and the LIP laboratory, focusing on parallel computation and algorithmic efficiency.

David DefourView profile

Professor

• Computer Arithmetic
• High Performance Computing
• Computer Architecture
• +11 more

David Defour is a Professor in Computer Science at the University of Perpignan, affiliated with the Faculty of Science and Engineering and the Mathematics and Computer Science Department. He conducts research at the Multidisciplinary Modeling and Simulation Laboratory (LAMPS) and serves as Scientific Coordinator of the MESO@LR Computing Center. His research focuses on the critical intersection of computer arithmetic and computer architecture, with emphasis on: Unconventional arithmetic systems (interval, fuzzy, high/reduced precision) Numerical reproducibility and error estimation in computational workflows Hardware-software co-design for GPUs, multicore systems, and embedded architectures Reverse engineering of computational units and power consumption optimization Defour actively mentors master's students through thesis proposals on numerical bug detection using LLMs, SAT solvers for numerical challenges, soft error resilience in floating-point units, and reverse execution debugging for HPC. He leads the ANR InterFlop project (2021-2025) and previously managed the BPS collaboration contract (2021-2022), while serving as IT expert for the High Court of Perpignan. The LAMPS laboratory under his research activities structures work around two core thematic axes: Characterization of Digital and Discrete Systems, and Mathematical and Numerical Modeling for Mechanics.

Fabienne JÉZÉQUELView profile

Lecturer

• Numerical algorithms
• High-performance computing
• Numerical validation

Fabienne JÉZÉQUEL is a Lecturer in Computer Science at Paris-Panthéon-Assas University, affiliated with the Sorbonne University Computer Science Laboratory (LIP6) and leading the PEQUAN (Performance and Quality of Digital Algorithms) team. Her research focuses on Numerical algorithms , High-performance computing , and Numerical validation , with applications in precision control and simulation reliability. Habilitation to Supervise Research (HDR) in Computer Science, University of Paris 6 (2005) Doctorate in Computer Science, University of Paris 6 (1996) Research Themes : Numerical algorithms with dynamic precision control High-performance computing (HPC) reliability and reproducibility Stochastic arithmetic for error estimation GPU acceleration of numerical methods Recent Publications : Focus on compensated algorithms, half-precision validation, and numerical reproducibility in extreme-scale simulations. Notable contributions include advancing stochastic arithmetic for trustworthy HPC codes and optimizing CADNA for GPU performance. Administrative Roles : Member of the office of the department of economic sciences at Paris-Panthéon-Assas University Member of the LIP6 Scientific Council

Ghigliotti GiovanniView profile

Lecturer

• Multiphase Fluid Dynamics
• Numerical Simulation
• Heat Transfer
• +3 more

Giovanni Ghigliotti is a Lecturer and researcher at the University of Grenoble Alpes since 2014, affiliated with the LEGI laboratory (UMR 5519). He is part of the MOST team, specializing in Turbulence Modeling and Simulation. His research focuses on multiphase fluid dynamics, particularly phase-change phenomena like boiling and cavitation, with an emphasis on numerical simulation to study hydrodynamic interactions, wetting, and material deformation. He holds a PhD in Fluid Mechanics from Joseph-Fourier University (2010). His work spans cavitation erosion mechanisms, fluid-structure interaction modeling, and heat transfer in multiphase systems. He contributes to the development of the Yales2 code, a parallel numerical simulation tool for complex flows. In teaching, he leads the Energy pathway of the Master's program in Process Engineering at Grenoble Alpes, instructing fluid mechanics, numerical methods, and thermodynamics. His research also involves collaborations with the GIS SUCCESS group, advancing high-performance computing and turbulence modeling. Key research themes include: Boiling crisis dynamics and thermal insulation effects Cavitation bubble collapse and material erosion Centrifugal microencapsulation process analysis Fluid-structure interaction in multiphase flows Unstructured grid simulations for phase change phenomena His recent publications (2024-2019) demonstrate expertise in non-Newtonian fluid simulation, cavitation erosion mechanics, and advanced computational methods for multiphase systems.

Yves RobertView profile

Professor

• Parallel Computing
• Scheduling Algorithms
• High-Performance Computing (HPC)
• +5 more

Yves Robert is a Professor at the Computer Science Laboratory LIP (Laboratoire de l'Informatique du Parallélisme) at École Normale Supérieure de Lyon (ENS Lyon). He also holds a Visiting Scientist position at the University of Tennessee Knoxville since 2011. His research focuses on resource optimization, parallel algorithms, scheduling techniques, and resilience methods in high-performance computing (HPC). Key areas include fault-tolerant numerical algorithms, energy-efficient scheduling, and iterative application checkpointing strategies. Robert has received prestigious awards such as the 2020 IEEE Computer Society Charles Babbage Award for contributions to parallel algorithms and scheduling. He is an IEEE Fellow (since 2005) and a Senior Member of Institut Universitaire de France. His work bridges theoretical foundations and practical algorithmic solutions, addressing challenges in distributed-memory architectures, workflow scheduling, and resilience at scale. His research spans parallelizing compilers, task graph scheduling, and energy-aware computing. Notable contributions include optimizing checkpointing strategies for HPC platforms, analyzing fault tolerance in iterative applications, and developing resilient scheduling heuristics for parallel jobs. His work often emphasizes scalability and reliability in modern computing systems.

• Image Processing
• Computer Science
• Mathematics
• +7 more

Carole Le Guyader is an Associate Professor at the Department of Mathematical and Software Engineering at the National Institute of Applied Sciences of Rouen (INSA Rouen). Her research focuses on hybrid variational/deep learning methods under topological and geometric constraints, with applications in biomedical imaging, geophysics, and high-performance computing (HPC). She supervises Ph.D. students and collaborates with institutions such as UCLA , Cambridge University , and University of Pau . Teaching: Mathematical Engineering (PDEs, CAGD in M1; Variational Methods in Image Processing in M2), L1/L2 STPI courses. Administrative Roles: LMI correspondent for GDR MIA (since 2021), Head of Mathematics Teaching at STPI INSA Rouen (since 2019), Deputy Director of GM Department (2018-2020). Her recent work emphasizes integrating deep learning with variational methods for tasks like semantic segmentation in medical imaging and nonlinear elasticity models for image registration. She has contributed to parallel computing implementations and industrial partnerships with Siemens-Gamesa and Dassault Systèmes .

Stephane VialleView profile

Researcher

• High-Performance Computing
• GPU Computing
• Quantum Computing
• +3 more

Stephane Vialle is a researcher at CentraleSupélec, leading the Interdisciplinary Laboratory of Digital Sciences. His research focuses on High-Performance Computing (HPC), GPU Computing, Quantum Computing, and Quantum Machine Learning. He has extensive experience in developing scalable fine-grained computing environments and optimizing parallel algorithms for distributed systems. His work spans financial engineering, energy management, and railway infrastructure through digital twin technology. Recent contributions include advancements in GPU cluster energy efficiency, stochastic control algorithms, and hybrid classical-quantum architectures for data clustering. Research interests emphasize optimizing parallel computing frameworks for diverse applications, including financial modeling, material science simulations, and transportation systems. His publications demonstrate expertise in distributed computing, fault-tolerant architectures, and algorithmic innovation across multiple computational paradigms. Current projects explore quantum computing integration with classical systems, GPU-based large-scale data processing, and real-world applications of parallel simulation techniques. Notable contributions include the parXXL development environment for coarse-grained platforms and the MINERVE digital twin for railway infrastructure management. His work bridges theoretical computing advancements with practical implementations in engineering and finance domains.

Vincent DanjeanView profile

Associate Professor

• Parallel Computing
• High-Performance Computing
• Task-Based Parallelism
• +6 more

Vincent Danjean is an associate professor at Grenoble Alpes University , specializing in parallel computing, high-performance computing, and bioinformatics. He earned his PhD in 2004 from École Normale Supérieure de Lyon under the supervision of Raymond Namyst. Research Interests: Vincent's work spans several critical areas in computational science: Parallel and Distributed Systems: Focus on task-based parallelism and hybrid cluster architectures. Performance Analysis: Development of visual frameworks for analyzing parallel applications. Bioinformatics: Application of computational methods to genetic and genomic data analysis. GPU Computing: Efficient scheduling and work stealing strategies for multi-GPU systems. Reproducible Research: Workflows using Git and Org-mode for scientific transparency. Publication Trends: His publications demonstrate a consistent focus on advancing parallel computing techniques, with significant contributions to GPU scheduling, cache-efficient algorithms, and visualization tools. Recent work includes interdisciplinary applications in genomics and cybersecurity protocols. Contact: vincent.danjean@imag.fr