Key concepts for parallel out-of-core LU factorization
Parallel Computing - Special double issue on environment and tools for parallel scientific computing
ScaLAPACK user's guide
A survey of out-of-core algorithms in numerical linear algebra
External memory algorithms
Journal of Parallel and Distributed Computing
Parallel Matrix Multiplication on a Linear Array with a Reconfigurable Pipelined Bus System
IEEE Transactions on Computers
Matrix Multiplication on Heterogeneous Platforms
IEEE Transactions on Parallel and Distributed Systems
A Proposal for a Heterogeneous Cluster ScaLAPACK (Dense Linear Solvers)
IEEE Transactions on Computers
Dynamic Matching and Scheduling of a Class of Independent Tasks onto Heterogeneous Computing Systems
HCW '99 Proceedings of the Eighth Heterogeneous Computing Workshop
I/O complexity: The red-blue pebble game
STOC '81 Proceedings of the thirteenth annual ACM symposium on Theory of computing
Efficient collective communication in distributed heterogeneous systems
Journal of Parallel and Distributed Computing
A cellular computer to implement the kalman filter algorithm
A cellular computer to implement the kalman filter algorithm
Scheduling Strategies for Master-Slave Tasking on Heterogeneous Processor Platforms
IEEE Transactions on Parallel and Distributed Systems
Communication lower bounds for distributed-memory matrix multiplication
Journal of Parallel and Distributed Computing
IEEE Transactions on Parallel and Distributed Systems
Algorithmic issues in grid computing
Algorithms and theory of computation handbook
Euro-Par'11 Proceedings of the 2011 international conference on Parallel Processing
| Hi-index | 0.00 |
This paper is focused on designing efficient parallel matrix-product algorithms for heterogeneous master-worker platforms. While matrix-product is well-understood for homogeneous 2D-arrays of processors (e.g., Cannon algorithm and ScaLAPACK outer product algorithm), there are three key hypotheses that render our work original and innovative: - Centralized data. We assume that all matrix files originate from, and must be returned to, the master. The master distributes data and computations to the workers while in ScaLAPACK, input and output matrices are supposed to be equally distributed among participating resources beforehand). Typically, our approach is useful in the context of speeding up MATLAB or SCILAB clients running on a server (which acts as the master and initial repository of files). - Heterogeneous star-shaped platforms. We target fully heterogeneous platforms, where computational resources have different computing powers. Also, the workers are connected to the master by links of different capacities. This framework is realistic when deploying the application from the server, which is responsible for enrolling authorized resources. - Limited memory. As we investigate the parallelization of large problems, we cannot assume that full matrix column blocks can be stored in the worker memories and be re-used for subsequent updates (as in ScaLAPACK). We have devised efficient algorithms for resource selection (deciding which workers to enroll) and communication ordering (both for input and result messages), and we report a set of numerical experiments on a platform at our site. The experiments show that our matrix-product algorithm has smaller execution times than existing ones, while it also uses fewer resources.