Partners

Main.Partners History

Hide minor edits - Show changes to markup

June 03, 2013, at 09:47 AM by 138.96.228.32 -
Changed lines 138-140 from:
to:
February 08, 2013, at 12:06 PM by 134.158.71.159 -
Added line 22:
February 08, 2013, at 12:06 PM by 134.158.71.159 -
Added line 32:
  • Augustin Degomme
Added line 54:
  • Jonathan Rouzaud-Cornabas
Changed lines 56-57 from:
to:
  • Pierre Veyre
Changed lines 81-82 from:
to:
  • Paul Renaud-Goud
Added line 111:
  • Takahiro Hirofushi
February 05, 2013, at 06:56 AM by 78.241.17.52 -
Added line 90:
  • Marc Frincu
March 16, 2012, at 01:29 PM by 134.158.32.169 -
Changed line 15 from:
  • The French National Research Agency (ANR) - contract No : ANR-11-INFRA-022
to:
  • The French National Research Agency (ANR) - contract No : ANR-11-INFRA-13
December 19, 2011, at 05:16 PM by 134.158.71.159 -
Changed lines 22-24 from:
  • M. Quinson (Coordinator)
to:
  • Lucas Nussbaum
  • Martin Quinson (Coordinator)
Changed lines 32-33 from:
to:
Changed lines 51-52 from:
to:
Changed lines 71-72 from:
to:
  • Denis Barthou
  • Olivier Beaumont
  • Nicolas Bonichon
  • Lionel Eyraud-Dubois
  • Brice Goglin
  • Abdou Guermouche
  • Fabien Mathieu
December 19, 2011, at 05:08 PM by 134.158.71.159 -
Changed line 1 from:

[[#Top]

to:

Added lines 71-77:
Changed lines 79-80 from:
to:
Added lines 85-93:
Changed lines 95-96 from:
to:
Added lines 100-116:
Changed lines 118-119 from:
to:
December 19, 2011, at 05:03 PM by 134.158.71.159 -
Added line 1:

[[#Top]

Added lines 25-26:
Added lines 34-35:
Added lines 50-51:
Changed lines 61-80 from:

The Runtime research project takes place within the

  context of high-performance computing (HPC). It seeks to explore the
  design, the implementation and the evaluation of novel mechanisms
  needed by runtime systems for parallel computers. Runtime systems
  are intermediate software layers providing parallel programming
  environments with specific functionalities left unaddressed by the
  underlying operating system. The increasing complexity of modern
  parallel hardware, making it more and more necessary to postpone
  essential decisions and actions (scheduling, optimizations) at run
  time, emphasizes the role of runtime systems. The research areas of
  the team center around three main directions: 1) Mastering large,
  hierarchical multiprocessor machines, 2) Optimizing communication
  over high performance clusters, and 3) Integrating Communications
  and Multithreading.

In the context of this project, HiePACS and RUNTIME teams are natural experts in the study and modelisation of HPC systems and applications. The CEPAGE team brings a strong expertise in Peer-to-Peer systems, in optimization for resource allocation, and in modelisation.

to:

The Runtime research project takes place within the context of high-performance computing (HPC). It seeks to explore the design, the implementation and the evaluation of novel mechanisms needed by runtime systems for parallel computers. Runtime systems are intermediate software layers providing parallel programming environments with specific functionalities left unaddressed by the underlying operating system. The increasing complexity of modern parallel hardware, making it more and more necessary to postpone essential decisions and actions (scheduling, optimizations) at runtime, emphasizes the role of runtime systems. The research areas of the team center around three main directions: 1) Mastering large, hierarchical multiprocessor machines, 2) Optimizing communication over high performance clusters, and 3) Integrating Communications and Multithreading.

In the context of this project, HiePACS and RUNTIME teams are natural experts in the study and modelisation of HPC systems and applications. The CEPAGE team brings a strong expertise in Peer-to-Peer systems, in optimization for resource allocation, and in modelisation.

Added lines 68-69:
Added lines 74-75:
Added lines 80-81:
Added lines 85-86:
December 19, 2011, at 05:00 PM by 134.158.71.159 -
Changed line 25 from:

MESCAL (Middleware Efficiently SCALable) is a joint team with members from CNRS, INRIA, Grenoble INP, and University Joseph Fourier, all of whom are in the LIG laboratory (UMR 5217). The goal of MESCAL is to design software solutions for the efficient exploitation of large distributed architectures at metropolitan, national and international scales. Their applications are intensive scientific computations. Their methodology is based on stochastic modelling of large discrete event systems, performance evaluation and simulation of large deterministic and probabilistic systems, middleware design, distributed systems. Hence, the skills and centers of interest of the MESCAL team are thus central to the whole \acronym project, which explains the important implication of the Grenoble site in all

to:

MESCAL (Middleware Efficiently SCALable) is a joint team with members from CNRS, INRIA, Grenoble INP, and University Joseph Fourier, all of whom are in the LIG laboratory (UMR 5217). The goal of MESCAL is to design software solutions for the efficient exploitation of large distributed architectures at metropolitan, national and international scales. Their applications are intensive scientific computations. Their methodology is based on stochastic modelling of large discrete event systems, performance evaluation and simulation of large deterministic and probabilistic systems, middleware design, distributed systems. Hence, the skills and centers of interest of the MESCAL team are thus central to the whole SONGS project, which explains the important implication of the Grenoble site in all

Added lines 46-73:

Three teams from INRIA Bordeaux -- Sud-Ouest are taking part to the project. They are joint with the PRES de Bordeaux and the CNRS (LaBRI UMR 5800).

CEPAGE aims at building strong foundations for distributed algorithms (graph exploration, black-hole search,...) and distributed data structures (routing, efficient query, compact labeling...) to understand how to explore large scale networks in the context of failures and how to disseminate data so as to answer quickly to specific queries. Secondly, it aims at building simple (based on local estimations without centralized knowledge), realistic models to represent accurately resource performance and to build a realistic view of the topology of the network (based on network coordinates, geometric spanners, δ-hyperbolic spaces). Then, it aims at proving that these models are tractable by providing low complexity distributed and randomized approximation algorithms for a set of basic scheduling problems (independent tasks scheduling, broadcasting, data dissemination,...) and associated overlay networks. At last, its goal is to prove the validity of the approach through softwares dedicated to several applications (molecular dynamics simulations, continuous integration) as well as more general tools.

The purpose of the HiePACS project is to efficiently perform frontier simulations arising from challenging research and industrial multiscale applications. The solution of these challenging problems requires a multidisciplinary approach involving applied mathematics, computational and computer sciences. In applied mathematics, it essentially involves advanced numerical schemes. In computational science, it involves massively parallel computing and the design of highly scalable algorithms and codes to be executed on future petaflop (and beyond) platforms. In order to tackle these large scale scientific challenges the HiePACS activities are organized as a continuum of research based on a multidisciplinary approach involving applied mathematics, computational and computer sciences. The research topics of HiePACS are the following: cover a wide area going from high performance computing on next generation architectures to the design of high performance solvers for linear algebra problems. Moreover, topics related to code coupling and high performance fast multipole methods are intensively studied. Finally, the team tries to take benefit from research results to improve the behavior of frontier simulations arising from challenging research and industrial multiscale applications.

The Runtime research project takes place within the

  context of high-performance computing (HPC). It seeks to explore the
  design, the implementation and the evaluation of novel mechanisms
  needed by runtime systems for parallel computers. Runtime systems
  are intermediate software layers providing parallel programming
  environments with specific functionalities left unaddressed by the
  underlying operating system. The increasing complexity of modern
  parallel hardware, making it more and more necessary to postpone
  essential decisions and actions (scheduling, optimizations) at run
  time, emphasizes the role of runtime systems. The research areas of
  the team center around three main directions: 1) Mastering large,
  hierarchical multiprocessor machines, 2) Optimizing communication
  over high performance clusters, and 3) Integrating Communications
  and Multithreading.

In the context of this project, HiePACS and RUNTIME teams are natural experts in the study and modelisation of HPC systems and applications. The CEPAGE team brings a strong expertise in Peer-to-Peer systems, in optimization for resource allocation, and in modelisation.

December 19, 2011, at 04:57 PM by 134.158.71.159 -
Added lines 25-27:

MESCAL (Middleware Efficiently SCALable) is a joint team with members from CNRS, INRIA, Grenoble INP, and University Joseph Fourier, all of whom are in the LIG laboratory (UMR 5217). The goal of MESCAL is to design software solutions for the efficient exploitation of large distributed architectures at metropolitan, national and international scales. Their applications are intensive scientific computations. Their methodology is based on stochastic modelling of large discrete event systems, performance evaluation and simulation of large deterministic and probabilistic systems, middleware design, distributed systems. Hence, the skills and centers of interest of the MESCAL team are thus central to the whole \acronym project, which explains the important implication of the Grenoble site in all simulation pillar tasks.

Added lines 32-41:
Added line 44:
December 19, 2011, at 04:16 PM by 134.158.71.159 -
Added lines 1-45:

Options: