Processor membership in asynchronous distributed systems
- 1 May 1994
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 5 (5) , 459-473
- https://doi.org/10.1109/71.282557
Abstract
Presents protocols for determining processor membership in asynchronous distributedsystems that are subject to processor and communication faults. These protocols dependon the placement of a total order on broadcast messages. The types of systems forwhich each of these protocols is applicable are characterized by the properties of thecommunication mechanisms and by the availability of stable storage. In the absence ofstable storage or of a mechanism for distinguishing promptly delivered messages, theauthors show that no membership protocol can exist. They also discuss their experiencein implementing these membership protocols.Keywords
This publication has 17 references indexed in Scilit:
- Necessary and sufficient conditions for broadcast consensus protocolsDistributed Computing, 1993
- Asynchronous Fault-Tolerant Total Ordering AlgorithmsSIAM Journal on Computing, 1993
- Trans: a reliable broadcast protocolIEE Proceedings I Communications, Speech and Vision, 1993
- Group communication in the Amoeba distributed operating systemPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1991
- Unreliable failure detectors for asynchronous systems (preliminary version)Published by Association for Computing Machinery (ACM) ,1991
- A fault-tolerant protocol for atomic broadcastIEEE Transactions on Parallel and Distributed Systems, 1990
- Broadcast protocols for distributed systemsIEEE Transactions on Parallel and Distributed Systems, 1990
- Preserving and using context information in interprocess communicationACM Transactions on Computer Systems, 1989
- Impossibility of distributed consensus with one faulty processJournal of the ACM, 1985
- Reliable broadcast protocolsACM Transactions on Computer Systems, 1984