Overview of the Project
REDIMPS (Research and Development of International Matrix Prediction System) is a project founded by the Strategic
Japanese-French Cooperative Program on
"Information and Communications Technology
including Computer Science" with the CNRS and the JST.
The goal of this international collaboration is to build
an international sparse linear equation solver expert site.
Among the objectives of
the project, one resides in the cooperation of the TLSE partners and the JAEA in the
testing, the validation and the promotion of the TLSE system that is
JAEA, who is one of the leading institute and organization of
Japanese HPC, is studying high-performance numerical simulation method
on novel super computers, and is expecting to find
the best linear solver within this collaboration.
By integrating knowledge and technology of JAEA
and TLSE partners, we aim at
promoting of an international expert system for sparse linear
algebra using an international grid computing environment.
- Test, validation and promotion of the TLSE system that is currently released.
- Interoperability between the
AEGIS/ITBL and DIET Grid Middleware is achieved.
- Annual France-Japan workshops on Grid computing and sparse linear algebra
are organized together with joined demontrations and presentations.
What's New or Scheduled
- Autumn 2009: Workshop held in Japan to finalize the
- June 2009: Venue of Dr Suzuki in Lyon
- April 2009: Meeting at the JAEA institute in Ueno, Tokyo,
- November 2008: Joined presentations and demonstrations held at SC'08
(Austin, USA) both in the JAEA and INRIA boothes.
REDIMPS Workshop, Tokyo, May 28-29 2007
- Venue of Dr Nishida in Toulouse and Lyon in december 2007
Joined demonstration at SC'07 in the JAEA booth at Reno (USA)
- Venue of Dr Caniou in Tokyo in Apr-May 2008
- The Second REDIMPS
Workshop was held at Toulouse, June 24-27 2008,
during the VECPAR'2008
International Conference on Computational Science within a special session
devoted to the Project.
- Venue of Dr Kushida in Lyon in June 2008
- Venue of Dr Caniou in Tokyo in July-August 2008. Exposé on DIET and Data Management
- Venue of Dr Caniou in Tokyo in september 2008
- Demonstration of the interoperability GridTLSE system in the
INRIA booth, and joined presentation in the JAEA booth in november 08
at SC'08 in Austin, Texas (USA)
- Meeting in JAEA institute in Tokyo, April 21 2009.
- Yves Caniou and Jean-Sébastien Gay. Simbatch: an API for
simulating and predicting the performance of parallel resources
managed by batch systems, In IEEE Workshop on Secure, Trusted,
Manageable and Controllable Grid Services (SGS) in parallel of
EuroPar'08, 12 p., 2008
- Yves Caniou and Noriyuki Kushida and Naowa
Teshima. Implementing interoperability between the AEGIS and DIET
GridRPC middleware to build an International Sparse Linear Algebra
Expert System, In the Second IEEE International Conference on
Advanced Engineering Computing and Applications in Sciences (ADVCOMP
2008), 6 p., 2008
- Yves Caniou, Jean-Sébastien Gay, and Pierre Ramet. Tunable
parallel experiments in a GridRPC framework: application to linear
solvers. In VECPAR'08 International Meeting on High Performance
Computing for Computational Science, 2008
- Noriyuki Kushida, Yoshio Suzuki, Naoya Teshima, Norihiro Nakajima,
Yves Caniou, Michel Daydé, and Pierre Ramet. Toward an
International Sparse Linear Algebra Expert System by Interconnecting
the ITBL Computational Grid with the Grid-TLSE Platform. In
VECPAR'08 International Meeting on High Performance Computing for
Computational Science, 2008
- K. Takase, H. Yoshida, T. Misawa. Large-Scale Two-Phase Flow Simulations in Fuel Bundles of Advanced Light Water Reactors
- S. Yamada, M. Okumura, M. Machida. High Performance Computing for Eigenvalue Solver in Density-Matrix Renormalization Group Method: Parallelization of the Hamiltonian Matrix-vector Multiplication.
- Hrachya Astsatryan, Vladimir Sahakyan, Youri Shoukouryan, Michel Daydé, Aurélie Hurault, Marc Pantel, Eddy Caron. A Grid-Aware Web Interface with Advanced Service Trading for Linear Algebra Calculations.. Dans : International Meeting High Performance Computing for Computational Science (VECPAR 2008), Toulouse, 24/06/08-27/06/08, 2008
- Web expert site for sparse linear algebra : GRID-TLSE
- Graph Partitioning : SCOTCH
- Parallel Solvers :
- LEGO (ANR Calcul Intensif et Grille de Calcul 2006-2009, ANR-CICG05-11)
SOLSTICE: (ANR Calcul Intensif et Simulation 2007-2010, ANR-06-CIS6-010)
- ASTER project (ITER)