O
Olivier Beaumont
Researcher at French Institute for Research in Computer Science and Automation
Publications - 150
Citations - 2694
Olivier Beaumont is an academic researcher from French Institute for Research in Computer Science and Automation. The author has contributed to research in topics: Scheduling (computing) & Approximation algorithm. The author has an hindex of 27, co-authored 141 publications receiving 2527 citations. Previous affiliations of Olivier Beaumont include École normale supérieure de Lyon & Korea University.
Papers
More filters
Journal ArticleDOI
Scheduling strategies for master-slave tasking on heterogeneous processor platforms
TL;DR: It is shown that finding the optimal steady state can be solved using a linear programming approach and, thus, in polynomial time, and a theoretical comparison of the computing power of tree-based versus arbitrary platforms is provided.
Journal ArticleDOI
Matrix multiplication on heterogeneous platforms
TL;DR: This work addresses the issue of implementing matrix multiplication on heterogeneous platforms with a (polynomial) column-based heuristic, which turns out to be very satisfactory and derives a theoretical performance guarantee and assesses its practical usefulness through MPI experiments.
Journal ArticleDOI
Scheduling divisible loads on star and tree networks: results and open problems
TL;DR: A unified theoretical perspective is proposed that synthesizes previously published results, several novel results, and open questions, in a view to foster hover divisible load scheduling research, and discusses both one-round and multiround algorithms.
Proceedings ArticleDOI
Bandwidth-centric allocation of independent tasks on heterogeneous platforms
TL;DR: A tree is used to model a grid, where resources can have different speeds of computation and communication, as well as different overlap capabilities, and it is shown how to determine the maximum steady-state throughput of a node in the base model, assuming the authors already know the throughput of the subtrees rooted at the node's children.
Journal ArticleDOI
A proposal for a heterogeneous cluster ScaLAPACK (dense linear solvers)
TL;DR: In this paper, the authors study the load balancing problem for dense linear algebra kernels on heterogeneous networks of workstations and propose a data allocation heuristic to balance the load on heterogenous platforms with respect to the performance of processors.