Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

An asynchronous algorithm for scattering information between the active nodes of a multicomputer system

Journal Article · · J. Parallel Distrib. Comput.; (United States)

In this paper the authors present an asynchronous algorithm for scattering information between the active nodes of a multicomputer system, having a large number of independent computers and workstations that are interconnected by a local area communication network. This algorithm is useful when it is desired to reduce the number of messages and the time delay necessary to transmit information from any node to all the active nodes of the system. The algorithm that the authors develop is based on one-way messages which are sent by each node, every unit of time, to a randomly selected node. The main advantage of this routing is that it overcomes inactive or faulty machines. They show that for an N node multicomputer in which n nodes are active, it is possible to scatter information to all the active nodes in approximately (1.693 + 1.414(1 - n/N))log/sub 2/n steps.

Research Organization:
School of Business and Economics, California State Univ., Fullerton, CA
OSTI ID:
6574637
Journal Information:
J. Parallel Distrib. Comput.; (United States), Journal Name: J. Parallel Distrib. Comput.; (United States) Vol. 3:3; ISSN JPDCE
Country of Publication:
United States
Language:
English

Similar Records

Method for securing concurrence among asynchronous processors
Journal Article · Sun Aug 01 00:00:00 EDT 1982 · IBM Tech. Disclosure Bull.; (United States) · OSTI ID:5214249

Adaptive fault-tolerant routing in hypercube multicomputers
Journal Article · Fri Nov 30 23:00:00 EST 1990 · IEEE Transactions on Computers (Institute of Electrical and Electronics Engineers); (USA) · OSTI ID:5862333

Authenticated algorithms for Byzantine agreement
Journal Article · Mon Oct 31 23:00:00 EST 1983 · SIAM J. Comput.; (United States) · OSTI ID:5170704