Sites Inria

Version française

PARIS Research team

Programming distributed parallel systems for large scale numerical simulation

  • Leader : Thierry Priol
  • Research center(s) : CRI Rennes - Bretagne Atlantique
  • Field : Networks, Systems and Services, Distributed Computing
  • Theme : Distributed and High Performance Computing
  • Partner(s) : Université Rennes 1,Ecole normale supérieure de Cachan,CNRS,Institut national des sciences appliquées de Rennes

Team presentation

The PARIS project-team aims at contributing to the programming of parallel and distributed systems for large scale numerical simulation applications. Such applications allow to speed-up the design of complex manufactured products, such as cars or aircraft, thanks to numerical simulation techniques. With the performance of computers rapidly increasing, it is possible to foresee in the near future comprehensive simulations of these designs that encompass multi-disciplinary aspects (structural mechanics, computational fluid dynamics, electromagnetism, noise analysis, etc). Numerical simulation of these different aspects will not be carried out by a single computer due to the lack of computing and memory resources. Instead, several computers and supercomputers will have to be used to keep simulation times within a reasonable bound. Moreover, simulation will have to be performed by different teams within a company or by different companies having appropriate skills and computing resources. These constraints advocate the use of distributed numerical simulation techniques. A distributed software infrastructure will therefore be required to connect all these simulation codes into a single environment that will have to provide basic services to support their execution. The PARIS project-team is carried out research activities in mainly three areas to reach its objective, the building of distributed software environments for large scale numerical simulation applications.

Research themes

  • Resource management for clusters
    Design of operating system mechanisms for resource management in clusters with a global approach to manage memory, disk, processor resources.
  • Run-times for parallel languages
    A Run-time is an interface between the compilers and the operating systems to support efficiently the execution of parallel programs. We are studying Run-times for several parallel languages: HPF, OpenMP and Java.
  • Software tools for distributed numerical simulation
    This research topic aims at contributing to the development of distributed software infrastructures to handle distributed numerical simulation applications.

International and industrial relations

  • Collaboration NEC-INRIA.
  • Collaboration with Aerospatiale-Matra
  • Participation to the Esprit HPCN R&D Jaco3 project

Keywords: Parallel programming Distributed programming Operating system Operating system Resource management Shared virtual memory Parallel file system Fault tolerance Cluster of pcs Distributed numerical simul