Methods for partitioning data to improve parallel execution time for sorting on heterogeneous clusters

Authors:
Christophe Cérin;Jean-Christophe Dubacq;Jean-Louis Roch
Affiliations:
LIPN, CNRS UMR 7030, Université de Paris Nord, Villetaneuse, France;LIPN, CNRS UMR 7030, Université de Paris Nord, Villetaneuse, France;CNRS – INRIA – INPG – UJF, Projet MOAIS, ID-IMAG, Montbonnot-Saint-Martin, France
Venue:
GPC'06 Proceedings of the First international conference on Advances in Grid and Pervasive Computing
Year:
2006

Citing 10
Cited 1

A logarithmic time sort for linear size networks

Journal of the ACM (JACM)
Parallel sorting by regular sampling

Journal of Parallel and Distributed Computing
On the versatility of parallel sorting by regular sampling

Parallel Computing
Parallel sorting by over partitioning

SPAA '94 Proceedings of the sixth annual ACM symposium on Parallel algorithms and architectures
A sequence of series for the Lambert W function

ISSAC '97 Proceedings of the 1997 international symposium on Symbolic and algebraic computation
Parallel Sorting Algorithms with Sampling Techniques on Clusters with Processors Running at Different Speeds

HiPC '00 Proceedings of the 7th International Conference on High Performance Computing
An Out-of-Core Sorting Algorithm for Clusters with Processors at Different Speed

IPDPS '02 Proceedings of the 16th International Parallel and Distributed Processing Symposium
A logarithmic time sort for linear size networks

STOC '83 Proceedings of the fifteenth annual ACM symposium on Theory of computing
Improving Parallel Execution Time of Sorting on Heterogeneous Clusters

SBAC-PAD '04 Proceedings of the 16th Symposium on Computer Architecture and High Performance Computing
On optimum multi-installment divisible load processing in heterogeneous distributed systems

Euro-Par'05 Proceedings of the 11th international Euro-Par conference on Parallel Processing

QCG-OMPI: MPI applications on grids

Future Generation Computer Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

The aim of the paper is to introduce general techniques in order to optimize the parallel execution time of sorting on a distributed architectures with processors of various speeds. Such an application requires a partitioning step. For uniformly related processors (processors speeds are related by a constant factor), we develop a constant time technique for mastering processor load and execution time in an heterogeneous environment and also a technique to deal with unknown cost functions. For non uniformly related processors, we use a technique based on dynamic programming. Most of the time, the solutions are in ${\mathcal O}$(p) (p is the number of processors), independent of the problem size n. Consequently, there is a small overhead regarding the problem we deal with but it is inherently limited by the knowing of time complexity of the portion of code following the partitioning.