DocumentCode :
1883634
Title :
Designing and Reconfiguring Fault-Tolerant Hypercubes
Author :
Farrag, Abdel Aziz ; Lou, Shituo
Author_Institution :
Dalhousie University, Canada
fYear :
2006
fDate :
14-17 May 2006
Firstpage :
11
Lastpage :
11
Abstract :
Several interconnection networks (such as meshes and hypercubes) can be modeled as circulant graphs. As a result, methods previously developed for constructing fault-tolerant solutions of circulant graphs can also be applied to these networks. Among these methods, the one based on the idea of "offsets partitioning" is the most efficient (for circulant graphs). We review this method in this paper, and extend its applications to hypercubes. Moreover, we develop new algorithms to reconfigure circulant graphs and hypercubes. Our results show that the fault-tolerant solutions obtained, and the reconfiguration algorithms developed are efficient.
Keywords :
Algorithm design and analysis; Application software; Computer networks; Computer science; Fault tolerance; Hypercubes; International collaboration; Multiprocessing systems; Multiprocessor interconnection networks; Partitioning algorithms;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
High-Performance Computing in an Advanced Collaborative Environment, 2006. HPCS 2006. 20th International Symposium on
ISSN :
1550-5243
Print_ISBN :
0-7695-2582-2
Type :
conf
DOI :
10.1109/HPCS.2006.19
Filename :
1628202
Link To Document :
بازگشت