Support for Fine-Grained Message Passing in Shared Memory Multiprocessors

From Tetherless World Wiki

Jump to: navigation, search

Citation: G. Byrd and Bruce A. Delagi. (1989) Support for Fine-Grained Message Passing in Shared Memory Multiprocessors. In KSL-89-15, 1989.

Publication techreport ( Edit )
type Technical Report
bibtype techreport
Bibtex basics
author G. Byrd and Bruce A. Delagi
title Support for Fine-Grained Message Passing in Shared Memory Multiprocessors
number KSL-89-15
institution Knowledge Systems, AI Laboratory
year 1989
Bibtex more
Access Paper
abstract Recent high-performance multiprocessors exploit cut-through routing, with packets routed as their first bytes arrive. Hardware-supported multicast can benefit the many parallel programs in which producers provide each value to multiple consumers. We describe several cut-through multicast protocols, including a restrictive (yet adaptive) routing scheme for deadlock avoidance. Simulations using synthetic and application-driven loads show it has significantly better performance than either multicast emulation or deadlock detection and resolution. The scheme provides cut-through multicast without requiring dedicated storage in the communication facilities for a full packet. We thus extend ideas considered for efficient cut-through routing in multiprocessor systems to include multicast.

KSL Technical Report ID: KSL-89-15
Facts about Support for Fine-Grained Message Passing in Shared Memory MultiprocessorsRDF feed
Abstract Recent high-performance multiprocessors ex Recent high-performance multiprocessors exploit cut-through routing, with packets routed as their first bytes arrive. Hardware-supported multicast can benefit the many parallel programs in which producers provide each value to multiple consumers. We describe several cut-through multicast protocols, including a restrictive (yet adaptive) routing scheme for deadlock avoidance. Simulations using synthetic and application-driven loads show it has significantly better performance than either multicast emulation or deadlock detection and resolution. The scheme provides cut-through multicast without requiring dedicated storage in the communication facilities for a full packet. We thus extend ideas considered for efficient cut-through routing in multiprocessor systems to include multicast. ltiprocessor systems to include multicast.
Author G. Byrd and Bruce A. Delagi  +
Bibtype techreport  +
Has author G. Byrd and Bruce A. Delagi  +
Has identifier KSL-89-15  +
Has publishing details 1989  +
Has title Support for Fine-Grained Message Passing in Shared Memory Multiprocessors  +
Has where published KSL-89-15  +
Has year 1989  +
Institution Knowledge Systems, AI Laboratory  +
Ksl tr id KSL-89-15  +
Number KSL-89-15  +
Process note YES  +
Title Support for Fine-Grained Message Passing in Shared Memory Multiprocessors  +
Year 1989  +