A5108188662- Parallel Computing and Optimization Techniques
- Distributed and Parallel Computing Systems
- Cloud Computing and Resource Management
- semigroups and automata theory
- Advanced Data Storage Technologies
- Advanced Algebra and Logic
- Embedded Systems Design Techniques
- Modular Robots and Swarm Intelligence
- Real-Time Systems Scheduling
- Distributed systems and fault tolerance
- Coding theory and cryptography
- Robotic Path Planning Algorithms
- Petri Nets in System Modeling
- Graph Theory and Algorithms
- Rough Sets and Fuzzy Logic
- Scientific Computing and Data Management
- Teaching and Learning Programming
- Geometric and Algebraic Topology
- Formal Methods in Verification
- Cryptography and Data Security
- Privacy-Preserving Technologies in Data
- Model-Driven Software Engineering Techniques
- Logic, programming, and type systems
- Business Process Modeling and Analysis
Laboratoire Bordelais de Recherche en Informatique
2007-2024
Université de Bordeaux
2002-2024
Centre National de la Recherche Scientifique
2024
Institut Polytechnique de Bordeaux
2002-2023
Centre de Recherche Inria Bordeaux - Sud-Ouest
2009-2023
Institut national de recherche en informatique et en automatique
2007-2018
Alternative Opportunities
2005-2014
Université de Lille
1991-1993
Abstract In the field of HPC, current hardware trend is to design multiprocessor architectures featuring heterogeneous technologies such as specialized coprocessors (e.g. Cell/BE) or data‐parallel accelerators GPUs). Approaching theoretical performance these a complex issue. Indeed, substantial efforts have already been devoted efficiently offload parts computations. However, designing an execution model that unifies all computing units and associated embedded memory remains main challenge....
Enabling HPC applications to perform efficiently when invoking multiple parallel libraries simultaneously is a great challenge. Even if uniform runtime system used underneath, scheduling tasks or threads coming from different over the same set of hardware resources introduces many issues, such as resource oversubscription, undesirable cache flushes and memory bus contention. This paper presents an extension StarPU, specifically designed for heterogeneous architectures, that allows codes run...
This paper presents EASYPAP, an easy-to-use programming environment designed to help students learn parallel programming. EASYPAP features a wide range of 2D computation kernels that the are invited parallelize using Pthreads, OpenMP, OpenCL or MPI. Execution can be interactively visualized, and powerful monitoring tools allow observe both scheduling computations assignment tiles threads/processes. By focusing on algorithms data distribution, experiment with diverse code variants tune...
The presentation of Peachy Parallel Assign-ments in several workshops on parallel and distributed computing education aims to promote the reuse high-quality assignments, both saving precious faculty time improving quality course assignments. Presented assignments are selected competitively- they must have been successfully used a real classroom, be easy for other instructors adopt, "cool inspirational" encourage students spend them talk about with others. Winning also archived Assignments...