Optimal Algorithms for Scheduling Large-Scale Divisible Load on Heterogeneous Systems in Non-blocking Mode of Communication

Authors:
Shang Mingsheng;Sun Shixin
Affiliations:
University of Electronic Science and Technology, P. R. China;University of Electronic Science and Technology, P. R. China
Venue:
HPCASIA '05 Proceedings of the Eighth International Conference on High-Performance Computing in Asia-Pacific Region
Year:
2005

Citing 8
Cited 0

Techniques for mapping tasks to machines in heterogeneous computing systems

Journal of Systems Architecture: the EUROMICRO Journal - Heterogeneous distributed and parallel architectures: hardware, software and design tools
On the Influence of Start-Up Costs in Scheduling Divisible Loads on Bus Networks

IEEE Transactions on Parallel and Distributed Systems
LoGPC: Modeling Network Contention in Message-Passing Programs

IEEE Transactions on Parallel and Distributed Systems
Ten Reasons to Use Divisible Load Theory

Computer
Efficient collective communication in distributed heterogeneous systems

Journal of Parallel and Distributed Computing
Divisible Load Theory: A New Paradigm for Load Scheduling in Distributed Systems

Cluster Computing
A Novel Optimal Load Distribution Algorithm for Divisible Loads

Cluster Computing
Scheduling Divisible Loads on Star and Tree Networks: Results and Open Problems

IEEE Transactions on Parallel and Distributed Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper considers scheduling large-scale divisible load on heterogeneous single-level tree networks to minimize the whole processing time. A more realistic and general model, i.e., both processors and links may have different speeds and arbitrary start-up costs, and with non-blocking mode of communication, is introduced. Under such environment, results are obtained including:(1) closedform expressions both for the processing time and the fraction of load for each processor are derived; (2) the influence of start-up costs on the optimal processing time is analyzed; and (3) for bounded number of processors with large-scale workload, optimal sequence and algorithms for distributing divisible load are proposed. Numerical examples are presented to illustrate the analysis.