论文信息 - Priority-based totally ordered multicast.

Priority-based totally ordered multicast.

The replicated state-machine approach is a general paradigm to implement fault-tolerant services that is particularly useful in real-time control applications. A totally ordered multicast protocol is a well-known method to enforce replica determinism in this approach. The paper presents an algorithm to provide a totally ordered multicast delivery service that takes priorities into account. The algorithm enforces the inter-replica coordination required to guarantee that high priority messages can be delivered before queued low priority messages that have not been delivered. The algorithm as been implemented as a variant of a protocol designed for local-area networks.

[1] Fred B. Schneider,et al. Replication management using the state-machine approach , 1993 .

[2] Leslie Lamport,et al. Time, clocks, and the ordering of events in a distributed system , 1978, CACM.

[3] Henri E. Bal,et al. An efficient reliable broadcast protocol , 1989, OPSR.

[4] Jo-Mei Chang,et al. Reliable broadcast protocols , 1984, TOCS.

[5] Paulo Veríssimo,et al. Formal Specification and Verification of a Network Independent Atomic Multicast Protocol , 1990, FORTE.

[6] André Schiper,et al. Virtually-synchronous communication based on a weak failure suspector , 1993, FTCS-23 The Twenty-Third International Symposium on Fault-Tolerant Computing.

[7] Richard D. Schlichting,et al. Preserving and using context information in interprocess communication , 1989, TOCS.

[8] Kenneth P. Birman,et al. Reliable communication in the presence of failures , 1987, TOCS.

[9] André Schiper,et al. Lightweight causal and atomic group multicast , 1991, TOCS.

[10] Paulo Veríssimo,et al. xAMp: a multi-primitive group communications service , 1992, [1992] Proceedings 11th Symposium on Reliable Distributed Systems.

[11] Liuba Shrira,et al. Lazy replication: exploiting the semantics of distributed services , 1990, ACM SIGOPS European Workshop.

[12] Danny Dolev,et al. Early delivery totally ordered multicast in asynchronous environments , 1993, FTCS-23 The Twenty-Third International Symposium on Fault-Tolerant Computing.