TY - GEN
T1 - Barrier synchronization in distributed-memory multiprocessors using rendezvous primitives
AU - Gupta, S. K.S.
AU - Panda, D. K.
N1 - Funding Information:
This research was supported in part by the Air Force Office of Scientific Research under grant number AFOSR90-0144, the NASA under grant number NAG-5-1987 and the NSF under grants number MIP-9106949 and number MIP-9205737
Publisher Copyright:
© 1993 IEEE.
Copyright:
Copyright 2019 Elsevier B.V., All rights reserved.
PY - 1993
Y1 - 1993
N2 - This paper deals with barrier synchronization in wormhole routed distributed-memory multiprocessors. New rendezvous and multirendezvous synchronization primitives are proposed to implement a barrier between two and multiple processors, respectively. These primitives reduce the number of communication steps required to implement a barrier; thus, significantly reducing the synchronization overhead for networks with high communication start-up cost. Two algorithms for barrier synchronization on k-ary n-cube networks are presented. The rendezvous primitive allows one to synchronize all processors in nlog2(k) steps. The multirendezvous primitive allows one to synchronize an arbitrary subset of processors in optimal number of communication steps depending on the ratio of the communication start-up (ts) to the link-propagation (tp) cost.
AB - This paper deals with barrier synchronization in wormhole routed distributed-memory multiprocessors. New rendezvous and multirendezvous synchronization primitives are proposed to implement a barrier between two and multiple processors, respectively. These primitives reduce the number of communication steps required to implement a barrier; thus, significantly reducing the synchronization overhead for networks with high communication start-up cost. Two algorithms for barrier synchronization on k-ary n-cube networks are presented. The rendezvous primitive allows one to synchronize all processors in nlog2(k) steps. The multirendezvous primitive allows one to synchronize an arbitrary subset of processors in optimal number of communication steps depending on the ratio of the communication start-up (ts) to the link-propagation (tp) cost.
UR - http://www.scopus.com/inward/record.url?scp=84858065246&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84858065246&partnerID=8YFLogxK
U2 - 10.1109/IPPS.1993.262826
DO - 10.1109/IPPS.1993.262826
M3 - Conference contribution
AN - SCOPUS:84858065246
T3 - Proceedings of 7th International Parallel Processing Symposium, IPPS 1993
SP - 501
EP - 505
BT - Proceedings of 7th International Parallel Processing Symposium, IPPS 1993
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 7th International Parallel Processing Symposium, IPPS 1993
Y2 - 13 April 1993 through 16 April 1993
ER -