Quorum-Based Replication in Asynchronous Crash-Recovery Distributed Systems (Research Note)
Euro-Par '00 Proceedings from the 6th International Euro-Par Conference on Parallel Processing
Group Communication Protocol for Multimedia Applications
ICCNMC '01 Proceedings of the 2001 International Conference on Computer Networks and Mobile Computing (ICCNMC'01)
IEEE Transactions on Knowledge and Data Engineering
Dealing efficiently with data-center disasters
Journal of Parallel and Distributed Computing
Total order broadcast and multicast algorithms: Taxonomy and survey
ACM Computing Surveys (CSUR)
Pronto: High availability for standard off-the-shelf databases
Journal of Parallel and Distributed Computing
High-speed group communication protocol for exchanging real-time multimedia data
International Journal of Wireless and Mobile Computing
International Journal of High Performance Computing and Networking
Heterogeneous clock group protocol for causally ordered delivery of messages
International Journal of Wireless and Mobile Computing
Fault tolerant and robust mutual exclusion protocol for synchronous distributed systems
ISCIS'05 Proceedings of the 20th international conference on Computer and Information Sciences
Causally ordered delivery of multimedia objects
Computer Communications
Avoiding disruptive failovers in transaction processing systems with multiple active nodes
Journal of Parallel and Distributed Computing
Hi-index | 0.00 |
Atomic Broadcast is a fundamental problem of distributed systems: it states that messages must be delivered in the same order to their destination processes. This paper describes a solution to this problem in asynchronous distributed systems in which processes can crash and recover.Chandra and Toueg have designed a Consensus-based solution to Atomic Broadcast problem for asynchronous distributed systems where crashed processes do not recover. Although our solution is based on different algorithmic principles, it follows the same approach: it transforms any Consensus protocol suited to the crash-recovery model into an Atomic Broadcast protocol suited to the same model. We show that Atomic Broadcast can be implemented without requiring any additional log operations in excess of those required by the Consensus. The paper also discusses how additional log operations can improve the protocol in terms of faster recovery and better throughput.