Two-layered protocol for a large-scale group of processes

A group including a larger number of processes implies larger computation and communication overheads to manipulate and transmit messages. We discuss a group which is composed of subgroups of processes to reduce the overheads. Each subgroup has a gateway process which communicates with the other gateway processes. We propose a protocol to causally deliver messages to processes in a group by using a vector of message sequence numbers whose size is the number of subgroups, smaller than number of processes. We evaluate the protocol.

[1]  Michel Raynal,et al.  Atomic broadcast in asynchronous crash-recovery distributed systems , 2000, Proceedings 20th IEEE International Conference on Distributed Computing Systems.

[2]  Louise E. Moser,et al.  Extended virtual synchrony , 1994, 14th International Conference on Distributed Computing Systems.

[3]  Makoto Takizawa,et al.  Delta-Causality and -delivery for wide-area group communications , 2000, Comput. Commun..

[4]  Akihito Nakamura,et al.  Causally ordering broadcast protocol , 1994, 14th International Conference on Distributed Computing Systems.

[5]  Roberto BALDONI,et al.  Efficient causally ordered communications for multimedia real-time applications , 1995, Proceedings of the Fourth IEEE International Symposium on High Performance Distributed Computing.

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

[7]  Andrew S. Tanenbaum,et al.  An evaluation of the Amoeba group communication system , 1996, Proceedings of 16th International Conference on Distributed Computing Systems.

[8]  Akihito Nakamura,et al.  Reliable broadcast protocol for selectively partially ordering PDUs (SPO protocol) , 1991, [1991] Proceedings. 11th International Conference on Distributed Computing Systems.

[9]  Georg Carle,et al.  Multicast Communication In Large Scale Networks , 1995, Third IEEE Workshop on the Architecture and Implementation of High Performance Communication Subsystems.

[10]  Friedemann Mattern,et al.  Virtual Time and Global States of Distributed Systems , 2002 .

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

[12]  Stephen E. Deering,et al.  Host groups: A multicast extension to the Internet Protocol , 1985, RFC.

[13]  Jon Postel,et al.  User Datagram Protocol , 1980, RFC.

[14]  Makoto Takizawa,et al.  Group communication protocol for real-time applications , 1998, Proceedings. 18th International Conference on Distributed Computing Systems (Cat. No.98CB36183).

[15]  Akihito Nakamura,et al.  Group communication protocol for large group , 1993, 1993 18th Conference on Local Computer Networks.

[16]  Kenneth P. Birman,et al.  Providing efficient, robust error recovery through randomization , 2001, Proceedings 21st International Conference on Distributed Computing Systems Workshops.