Title :
Load Balancing on Semi-direct Product and Hierarchical Star Interconnection Networks
Author_Institution :
Sch. of Inf., Yunnan Univ. of Finance & Econ., Kunming, China
Abstract :
We suggest strategies for solving the problem of load balancing on semi-direct product and hierarchical star networks to enhance the stability and accelerate the iterating speed over some known algorithms whose convergence speed is slow when they are applied to two sorts of multistage network above because their small Laplace condition numbers, relative to those small scale networks. Several theoretical results are presented to prove the superiority of our algorithms. Concretely, our schemes need a low-cost spectrum computing and simple iterating procedure in conjunction with a better convergence property, compared with known those. Some numerical results of stimulation validate the reliability of our algorithm.
Keywords :
distributed processing; iterative methods; multistage interconnection networks; resource allocation; stability; Laplace condition number; distributed systems; hierarchical star interconnection network; large-scale parallel systems; load balancing; multistage network; semidirect product network; Computer networks; Convergence; Distributed computing; Hydrogen; Iterative algorithms; Large-scale systems; Load management; Multiprocessor interconnection networks; Optimized production technology; Tin; Laplacian spectrum; Load balancing; hierarchical star; semi-direct product;
Conference_Titel :
Computational Sciences and Optimization, 2009. CSO 2009. International Joint Conference on
Conference_Location :
Sanya, Hainan
Print_ISBN :
978-0-7695-3605-7
DOI :
10.1109/CSO.2009.186