The pebble crurching model for load balancing in concurrent hypercube ensembles

Authors:
J. Barhen;S. Gulati;S. S. Iyengar
Affiliations:
Jet Propulsion Laboratory, Caltech, 4800 Oak Grove Drive, Pasadena, California;Department of Computer Science, Louisana State University, Baton Rouge, LA;Department of Computer Science, Louisana State University, Baton Rouge, LA
Venue:
C3P Proceedings of the third conference on Hypercube concurrent computers and applications: Architecture, software, computer systems, and general issues - Volume 1
Year:
1988

Citing 5
Cited 1

Optimal static load balancing in distributed computer systems

Journal of the ACM (JACM)
A Distributed Drafting Algorithm for Load Balancing

IEEE Transactions on Software Engineering
The Gradient Model Load Balancing Method

IEEE Transactions on Software Engineering - Special issue on distributed systems
A comparison of receiver-initiated and sender-initiated adaptive load sharing (extended abstract)

SIGMETRICS '85 Proceedings of the 1985 ACM SIGMETRICS conference on Measurement and modeling of computer systems
Graphs and Hypergraphs

Graphs and Hypergraphs

Neural networks and dynamic complex systems

ANSS '89 Proceedings of the 22nd annual symposium on Simulation

Quantified Score

Hi-index	0.00

Visualization

Abstract

The successful development of fifth generation systems require enormous computational capability and flexibility necessitating the ability to achieve operational responses in hard real-time through optimal resource utilization. This entails dynamically balancing the computational load among all the processing nodes in the system. We propose a graph-theoretic, receiver-initiated, distributed protocol for dynamic load balancing protocol in large-scale hypercube ensembles. Using attributed hypergraphs as the primary data structure for constraint modeling and dynamic optimization, we consider systems running precedence-constrained heterogeneous tasks. Fault Tolerance is ensured by incorporating an integrity check for the decision nodes and their subsequent reelection if needed. Simulation studies are used to evaluate the performance of the algorithm.