Abstract
A set of personal computers is connected to form a ring structured parallel system: Each processor has access to its local memory and can exchange messages with its two ring neighbors. A branch-and-bound procedure is implemented in Pascal to run in parallel on the ring and solve the Travelling-Salesman-Problem. Heuristics are developed to maintain a priority queue in a distributed heap. The computing times and speedups for 25 random graphs obtained with up to 16 ring members are discussed.
Cite
CITATION STYLE
Vornberger, O. (1986). Implementing branch-and-bound in a ring of processors. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 237 LNCS, pp. 157–164). Springer Verlag. https://doi.org/10.1007/3-540-16811-7_166
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
