TY - JOUR
T1 - Self-stabilizing routing and related protocols
AU - Dolev, Shlomi
N1 - Funding Information:
1This work was supported by NSF Presidential Young Investigator Award CCR-91-58478, by the Israeli Ministry of Science and Arts Grant #6756196, and funds from the Texas A&M University College of Engineering. A preliminary version of this work was presented in the 7th Workshop on Distributed Algorithms, September 1993. 2E-mail: [email protected].
PY - 1997/5/1
Y1 - 1997/5/1
N2 - Aself-stabilizingsystem is a distributed system which can tolerateany numberandany typeof faults in the history. After the last fault occurs the system converges to alegitimate behavior. The self-stabilization property is very useful for systems in which processors may malfunction for a while and then recover. When there is a long enough period during which no processor malfunctions the system stabilizes.Dynamicdistributed systems are systems in which communication links and processors may fail and recover during normal operation. Such failures could cause partitioning of the system communication graph. The application of self-stabilizing protocols to dynamic systems is natural. Following the last topology change each connected component of the system stabilizes independently. We present self-stabilizing dynamic protocols for a variety of tasks including: routing, leader election, and topology update. For systems that support local broadcasts to neighbors in a single time unit the protocol for each of those tasks stabilizes in Θ(d) time, wheredis theactualdiameter of the system.
AB - Aself-stabilizingsystem is a distributed system which can tolerateany numberandany typeof faults in the history. After the last fault occurs the system converges to alegitimate behavior. The self-stabilization property is very useful for systems in which processors may malfunction for a while and then recover. When there is a long enough period during which no processor malfunctions the system stabilizes.Dynamicdistributed systems are systems in which communication links and processors may fail and recover during normal operation. Such failures could cause partitioning of the system communication graph. The application of self-stabilizing protocols to dynamic systems is natural. Following the last topology change each connected component of the system stabilizes independently. We present self-stabilizing dynamic protocols for a variety of tasks including: routing, leader election, and topology update. For systems that support local broadcasts to neighbors in a single time unit the protocol for each of those tasks stabilizes in Θ(d) time, wheredis theactualdiameter of the system.
UR - http://www.scopus.com/inward/record.url?scp=0031143006&partnerID=8YFLogxK
U2 - 10.1006/jpdc.1997.1317
DO - 10.1006/jpdc.1997.1317
M3 - Article
AN - SCOPUS:0031143006
SN - 0743-7315
VL - 42
SP - 122
EP - 127
JO - Journal of Parallel and Distributed Computing
JF - Journal of Parallel and Distributed Computing
IS - 2
ER -