Online task scheduling on heterogeneous clusters: an experimental study

Authors:
Einar M. R. Rosenvinge;Anne C. Elster;Cyril Banino
Affiliations:
Department of Computer and Information Science (idi), Norwegian University of Science and Technology (ntnu);Department of Computer and Information Science (idi), Norwegian University of Science and Technology (ntnu);Department of Computer and Information Science (idi), Norwegian University of Science and Technology (ntnu)
Venue:
PARA'04 Proceedings of the 7th international conference on Applied Parallel Computing: state of the Art in Scientific Computing
Year:
2004

Citing 6
Cited 1

Allocating Independent Subtasks on Parallel Processors

IEEE Transactions on Software Engineering
Guided self-scheduling: A practical scheduling scheme for parallel supercomputers

IEEE Transactions on Computers
Factoring: a method for scheduling parallel loops

Communications of the ACM
Load-sharing in heterogeneous systems via weighted factoring

Proceedings of the eighth annual ACM symposium on Parallel algorithms and architectures
Scheduling Divisible Loads in Parallel and Distributed Systems

Scheduling Divisible Loads in Parallel and Distributed Systems
Data Staging Effects in Wide Area Task Farming Applications

CCGRID '01 Proceedings of the 1st International Symposium on Cluster Computing and the Grid

A load balancing strategy for computations on large, read-only data sets

PARA'06 Proceedings of the 8th international conference on Applied parallel computing: state of the art in scientific computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper considers effcient task scheduling methods for applications on heterogeneous clusters. The Master/Worker paradigm is used, where the independent tasks are maintained by a master node which hands out batches of a variable amount of tasks to requesting worker nodes. The Monitor strategy is introduced and compared to other strategies suggested in the literature. Our online strategy is especially suitable for heterogeneous clusters with dynamic loads.