Author pages are created from data sourced from our academic publisher partnerships and public sources.
Share This Author
A review on the prediction of building energy consumption
Optimized Schwarz Methods without Overlap for the Helmholtz Equation
A variant of the Schwarz method which converges without overlap for the Helmholtz equation is studied, and it is shown that the key ingredients for such an algorithm are the transmission conditions, which lead to convergence of the algorithm in a finite number of steps.
Optimal convergence properties of the FETI domain decomposition method
In this paper an original variant of the FETI domain decomposition method is introduced for heterogeneous media. This method uses new absorbing interface conditions in place of the Neumann interface…
Introduction to Grid Computing
The authors present several grid middlewares adopted in scientific applications, explain the mechanisms behind well-known grid projects, and apply the Monte Carlo method and PDEs to industrial problems in finance and computational mechanics.
An optimized Schwarz method with two‐sided Robin transmission conditions for the Helmholtz equation
A new optimized Schwarz method without overlap in the 2d case is presented, which uses a different Robin condition for neighbouring subdomains at their common interface, and which is called two‐sided Robin condition.
Two-level domain decomposition methods with Lagrange multipliers for the fast iterative solution of acoustic scattering problems
A Hadoop MapReduce Performance Prediction Method
- Ge Song, Zide Meng, F. Huet, F. Magoulès, Lei Yu, Xuelian Lin
- Computer ScienceIEEE 10th International Conference on High…
- 13 November 2013
A simple framework to predict the performance of Hadoop jobs is proposed that makes some theoretical cost models more practical, and also well fits for the diversification of the jobs and clusters.
Iterative Methods for Sparse Linear Systems on Graphics Processing Unit
- Abal-Kassim Cheik Ahamed, F. Magoulès
- Computer ScienceIEEE 14th International Conference on High…
- 25 June 2012
This paper considers iterative Krylov subspace based iterative solvers to take advantage of massive parallelism of Graphics Processing Unit (GPU), and discusses data format and data structure for sparse matrices, and how to efficiently implement these solvers on the Nvidia's CUDA platform.
A non Overlapping Domain Decomposition Method for the Exterior Helmholtz Problem
- A. D. L. Bourdonnaye, C. Farhat, Antonini Macedo, F. Magoulès, F. Roux
- Computer Science
- 1 October 1997
An alternative domain decomposition algorithm that is better suited for the exterior Helmholtz problem is presented, in a formalism that can use either one or two Lagrange multiplier fields for solving the corresponding interface problem by a Krylov method.