In this paper, we address the issue of implementing matrix-matrix multiplication on heterogeneous platforms. We target two different classes of heterogeneous computing resources: heterogeneous networks of workstations, and collections of heterogeneous clusters. Intuitively, the problem is to load balance the work with different-speed resources while minimizing the communication volume. We formally state this problem and prove its NP-completeness. Next we introduce a (polynomial) column-based heuristic, which turns out to be very satisfactory: we derive a theoretical performance guarantee for the heuristic, and we assess its practical usefulness through MPI experiments.Dans ce rapport, nous nous intéressons au problè me de l'implémentation d...
Parallel computing on networks of workstations are intensively used in some application areas such a...
International audienceWe consider the problem of data allocation when performing matrix multiplicati...
This paper presents and analyzes two different strategies of heterogeneous distribution of computati...
In this paper, we address the issue of implementing matrix-matrix multiplication on heterogeneous pl...
International audienceIn this paper, we address the issue of implementing matrix-matrix multiplicati...
In this paper, we address the issue of imple-menting matrix-matrix multiplication on heteroge-neous ...
Proceedings of the 8th IEEE International Conference on Cluster Computing (Cluster 2006), October, 2...
Proceedings of: Third International Workshop on Sustainable Ultrascale Computing Systems (NESUS 2016...
In this paper, an adaptive matrix multiplication algorithm for dynamic heterogeneous environments is...
We present a new approach to utilizing all CPU cores and all GPUs on heterogeneous multicore and mul...
In this report, we consider a simple but important linear algebra kernel, matrix-matrix multiplicati...
In this document, we describe two strategies of distribution of computations that can be used to imp...
Matrix multiplication is taken as a test bed for parallel processing on heterogeneous networks of wo...
2012 IEEE 26th Parallel and Distributed Processing Symposium Workshops and PhD Forum (IPDPSW), Shang...
Matrix multiplication is one of the important operations in scientific and engineering application. ...
Parallel computing on networks of workstations are intensively used in some application areas such a...
International audienceWe consider the problem of data allocation when performing matrix multiplicati...
This paper presents and analyzes two different strategies of heterogeneous distribution of computati...
In this paper, we address the issue of implementing matrix-matrix multiplication on heterogeneous pl...
International audienceIn this paper, we address the issue of implementing matrix-matrix multiplicati...
In this paper, we address the issue of imple-menting matrix-matrix multiplication on heteroge-neous ...
Proceedings of the 8th IEEE International Conference on Cluster Computing (Cluster 2006), October, 2...
Proceedings of: Third International Workshop on Sustainable Ultrascale Computing Systems (NESUS 2016...
In this paper, an adaptive matrix multiplication algorithm for dynamic heterogeneous environments is...
We present a new approach to utilizing all CPU cores and all GPUs on heterogeneous multicore and mul...
In this report, we consider a simple but important linear algebra kernel, matrix-matrix multiplicati...
In this document, we describe two strategies of distribution of computations that can be used to imp...
Matrix multiplication is taken as a test bed for parallel processing on heterogeneous networks of wo...
2012 IEEE 26th Parallel and Distributed Processing Symposium Workshops and PhD Forum (IPDPSW), Shang...
Matrix multiplication is one of the important operations in scientific and engineering application. ...
Parallel computing on networks of workstations are intensively used in some application areas such a...
International audienceWe consider the problem of data allocation when performing matrix multiplicati...
This paper presents and analyzes two different strategies of heterogeneous distribution of computati...