Electrical & Computer Engineering Faculty Publications
Document Type
Article
Publication Date
6-8-2005
Abstract
Load balancing for parallel computations is modeled as a deterministic dynamic nonlinear time-delay system. This model accounts for the trade-off between using processor time/network bandwidth and the advantage of distributing the load evenly between the nodes to reduce overall processing time. A distributed closed-loop controller is presented to balance load dynamically at each node by using not only the local estimate of the queue size of other nodes, but also estimates of the number of tasks in transit. A discrete event simulation using OPNET Modeler is presented and compared with experimental data, and results indicate good agreement between the nonlinear time-delay model and the behaviors observed on a parallel computer network. Moreover, both simulations and experiments show a dramatic increase in performance obtained using the proposed closed-loop controller.
Publisher
IEEE
Publication Title
Proceedings of the 2005 American Control Conference
ISSN
0743-1619
First Page
2721
Last Page
2726
DOI
10.1109/ACC.2005.1470380
Language (ISO)
English
Sponsorship
IEEE
Keywords
Bandwidth, Chaotic communication, Computational modeling
Recommended Citation
Abdallah, Chaouki T.; Zhong Tang; John White; John Chiasson; J. Douglas Birdwell; and Majeed M. Hayat. "Closed-loop load balancing: Comparison of a discrete event simulation with experiments." Proceedings of the 2005 American Control Conference (2005): 2721-2726. doi:10.1109/ACC.2005.1470380.
