Processor allocation for a class of hypercube-like supercomputers
Document Type
Conference Proceeding
Publication Date
12-1-1992
Abstract
Modified hypercubes (MHs) [20] have been proposed as the building blocks of hypercube-based parallel systems that support incremental growth techniques. In contrast, systems comprising the standard hypercube can not be expanded in practice. Processor allocation for MHs is a more difficult task due to a slight deviation in their topology from that of the hypercube. This paper addresses the processor allocation problem for MHs and proposes two strategies which are based, partially or entirely, on a table look-up approach. The proposed strategies are characterized by a perfect subcube recognition ability and a superior performance. Further, two existing processor allocation strategies for pure hypercubes, namely the buddy and free list strategies, are shown to be ineffective for MHs, in the light of their inability to recognize many available subcubes.
Identifier
84943730067 (Scopus)
ISBN
[0818626305]
Publication Title
Proceedings of the International Conference on Supercomputing
External Full Text Location
https://doi.org/10.1109/superc.1992.236689
First Page
740
Last Page
749
Volume
Part F129723
Recommended Citation
Haravu, Nagasimha G. and Ziavras, Sotirios G., "Processor allocation for a class of hypercube-like supercomputers" (1992). Faculty Publications. 17270.
https://digitalcommons.njit.edu/fac_pubs/17270
