Processor allocation and communication in networks
Thesis/Dissertation
·
OSTI ID:6036610
The author will describe two known strategies for static processor allocation in an n-cube multiprocessor, namely the buddy system strategy and the gray code strategy, and then propose a new strategy that out-performs the first by (n-k+1) and the second by (n-k+1)/2 in cube recognition. Furthermore, his strategy is suitable for static as well as dynamic processor allocation and results in less system fragmentation, more subcubes recognition and higher fault tolerance. He also introduces an extension to strategy that will enhance the performance drastically so that his algorithm, together with the extension, will outperform the buddy system and the gray strategy. The cube-connected cycles network is an attractive alternative for the n-cube, since it overcomes the degree problem by limiting it to 3. He proposes and studies cube-connected cycles networks that are missing some of their cycles, called incomplete cube-connected cycles networks. These incomplete cube-connected cycles networks can be used to interconnect systems with any number of processors, thus overcoming the severe restriction of the complete cube-connected cycles networks. In incomplete cube-connected networks the processors are well utilized, routes are shorter and the layout area is a lot smaller. He developed efficient routing and broadcasting algorithms and shows how to lay this network on a chip. Broadcasting is a procedure by which a processor can pass a message to all other processors in the network nonredundantly. He briefly describes the algorithm for broadcasting in the hypercube developed by Sullivan and Bashkow, than he develops a simple yet efficient fault tolerant broadcasting algorithm for the hypercube in the presence of some faulty processors.
- Research Organization:
- Oregon State Univ., Corvallis, OR (USA)
- OSTI ID:
- 6036610
- Country of Publication:
- United States
- Language:
- English
Similar Records
Processor allocation in an N-cube multiprocessor using Gray codes
A top-down processor allocation scheme for hypercube computers
Subcube allocation and task migration in hypercube multiprocessors
Journal Article
·
Mon Nov 30 23:00:00 EST 1987
· IEEE Trans. Comput.; (United States)
·
OSTI ID:5384833
A top-down processor allocation scheme for hypercube computers
Journal Article
·
Mon Dec 31 23:00:00 EST 1990
· IEEE Transactions on Parallel and Distributed Systems (Institute of Electrical and Electronics Engineers); (United States)
·
OSTI ID:6098173
Subcube allocation and task migration in hypercube multiprocessors
Journal Article
·
Sat Sep 01 00:00:00 EDT 1990
· IEEE Transactions on Computers (Institute of Electrical and Electronics Engineers); (USA)
·
OSTI ID:6541206