As both resources and applications are becoming more complex, resource management also becomes a more challenging task. For example, scheduling code-coupling applications on federations of clusters such as Grids results in complex resource selection algorithms. The abstractions provided by current Resource Management Systems (RMS) - usually rigid jobs or advance reservations - are insufficient to enable such applications to efficiently select resources. This paper studies an RMS architecture that delegates resource selection to applications while the RMS still keeps control over the resources. The proposed architecture is evaluated using a simulator which is then validated with a proof-of-concept implementation. Results show that such a sys...
Stacking a multiprocessor (MPSoC) layer and a FPGA layer to form a 3D Recon gurable System-on-Chip (...
La résolution exacte de problèmes d'optimisation combinatoire de grande taille, tels que les problèm...
Most HPC platforms require users to submit a pre-determined number of computation requests (also cal...
As both resources and applications are becoming more complex, resource management also becomes a mor...
Non-predictably evolving applications are applications that change their resource requirements durin...
High-Performance Computing (HPC) resources, such as Supercomputers, Clusters, Grids and HPC Clouds, ...
National audienceThe race for ever more computing power raises the issue of supercomputers' power co...
Multi-core architectures using scratch pad memories are very attractive to execute embedded time-cri...
The work presented in this Thesis is about scheduling applications in computational Grids. We study ...
Community clouds have arisen as a promising infrastructure to face the ever-growing demand for compu...
Grid services often consist of remote sequential or rigid parallel application executions. However, ...
Task-based systems have gained popularity because of their promise of exploiting the computational p...
La programmation par contraintes est une technique puissante pour résoudre, entre autres, des problè...
Cloud Radio Access Network (C-RAN) has been proposed as a promising architecture to meet the exponen...
RÉSUMÉ: Les échéanciers sont très importants dans le contexte de la gestion de projet. Le problème d...
Stacking a multiprocessor (MPSoC) layer and a FPGA layer to form a 3D Recon gurable System-on-Chip (...
La résolution exacte de problèmes d'optimisation combinatoire de grande taille, tels que les problèm...
Most HPC platforms require users to submit a pre-determined number of computation requests (also cal...
As both resources and applications are becoming more complex, resource management also becomes a mor...
Non-predictably evolving applications are applications that change their resource requirements durin...
High-Performance Computing (HPC) resources, such as Supercomputers, Clusters, Grids and HPC Clouds, ...
National audienceThe race for ever more computing power raises the issue of supercomputers' power co...
Multi-core architectures using scratch pad memories are very attractive to execute embedded time-cri...
The work presented in this Thesis is about scheduling applications in computational Grids. We study ...
Community clouds have arisen as a promising infrastructure to face the ever-growing demand for compu...
Grid services often consist of remote sequential or rigid parallel application executions. However, ...
Task-based systems have gained popularity because of their promise of exploiting the computational p...
La programmation par contraintes est une technique puissante pour résoudre, entre autres, des problè...
Cloud Radio Access Network (C-RAN) has been proposed as a promising architecture to meet the exponen...
RÉSUMÉ: Les échéanciers sont très importants dans le contexte de la gestion de projet. Le problème d...
Stacking a multiprocessor (MPSoC) layer and a FPGA layer to form a 3D Recon gurable System-on-Chip (...
La résolution exacte de problèmes d'optimisation combinatoire de grande taille, tels que les problèm...
Most HPC platforms require users to submit a pre-determined number of computation requests (also cal...