Load management for load balancing on heterogeneous platforms: A comparison of traditional and neural network based approaches
In this paper we compare simple load metrics with neural networks which have been trained to predict the expected delay of an application from the sampled load informations. The results show that the proposed load metric performs well in heterogeneous environments. Further, neural networks can improve the performance of load balancing facilities.
Unable to display preview. Download preview PDF.
- 1.D. Ferrari and S. Zhou. An empirical investigation of load indices for load balancing applications. In Performance '87, pages 515–528. Elsevier Science Publishers, 1988.Google Scholar
- 2.S. Haykin. Neural Networks: A Comprehensive Foundation. Macmillian College Publishing Company, New York, 1994.Google Scholar
- 4.Pankaj Mehra. Automated Learning of Load-Balancing Strategies for a Distributed Computer System. PhD thesis, University of Illinois at Urbana-Champaign, 1993. Available on ftp.ibr.cs.tu-bs.de.Google Scholar
- 5.I. Miller and J.E. Freund. Probability and Statistics for Engineers. Prentice-Hall, Englewood Cliffs, 2nd edition, 1977.Google Scholar
- 6.B. Schnor, H. Langendörfer, and S. Petri. Einsatz neuronaler Netze zur Lastbalancierung in Workstationclustern. In Proc. Praxisorientierte Parallelverarbeitung, pages 154–165, Braunschweig, October 1994.Google Scholar
- 7.B. Schnor, S. Petri, and H. Langendörfer. Using neural networks for prediction of load indices in heterogeneous computing environments. In Preparation, 1996.Google Scholar