Unconscious Eventual Consistency with Gossips

This paper combines various self-stabilization techniques within a replication protocol that ensures eventual consistency in largescale distributed systems subject to network partitions and asynchrony. A simulation study shows that the resulting protocol is scalable and achieves high throughput under load. Our protocol does not rely on any form of consensus, which would lead to block the replicas in case of partitions and asynchrony. Our protocol instead ensures that (1) updates are continuously applied to the replicas and (2) no two updates are ever performed in a different order. Gaps might occur during periods of unreliable communication. They are filled whenever connectivity is provided, and consistency is then eventually ensured, but without any conscious commitment. That is, there is no point in the computation when replicas know that consistency is achieved. This unconsciousness is the key to tolerating perpetual asynchrony with no consensus support.

[1]  Miguel Castro,et al.  Practical byzantine fault tolerance and proactive recovery , 2002, TOCS.

[2]  Anne-Marie Kermarrec,et al.  Efficient epidemic-style protocols for reliable and scalable multicast , 2002, 21st IEEE Symposium on Reliable Distributed Systems, 2002. Proceedings..

[3]  Francisco Moura,et al.  Optimistic total order in wide area networks , 2002, 21st IEEE Symposium on Reliable Distributed Systems, 2002. Proceedings..

[4]  Mohamed G. Gouda,et al.  Stabilizing Communication Protocols , 1991, IEEE Trans. Computers.

[5]  Ben Y. Zhao,et al.  OceanStore: an architecture for global-scale persistent storage , 2000, SIGP.

[6]  Anne-Marie Kermarrec,et al.  Probabilistic Reliable Dissemination in Large-Scale Systems , 2003, IEEE Trans. Parallel Distributed Syst..

[7]  Marvin Theimer,et al.  Managing update conflicts in Bayou, a weakly connected replicated storage system , 1995, SOSP.

[8]  Magnus Karlsson,et al.  Taming aggressive replication in the Pangaea wide-area file system , 2002, OPSR.

[9]  Nancy A. Lynch,et al.  Eventually-Serializable Data Services , 1999, Theor. Comput. Sci..

[10]  Karl Aberer,et al.  Improving Data Access in P2P Systems , 2002, IEEE Internet Comput..

[11]  Kevin Curran,et al.  Online Gaming , 2005, EuroIMSA.

[12]  Nancy A. Lynch,et al.  Impossibility of distributed consensus with one faulty process , 1985, JACM.

[13]  Krishna P. Gummadi,et al.  Measurement, modeling, and analysis of a peer-to-peer file-sharing workload , 2003, SOSP '03.

[14]  Roman Schmidt,et al.  Technical University of Vienna Improving Data Access in P 2 P Systems , 2001 .

[15]  Luís E. T. Rodrigues,et al.  An indulgent uniform total order algorithm with optimistic delivery , 2002, 21st IEEE Symposium on Reliable Distributed Systems, 2002. Proceedings..

[16]  Anne-Marie Kermarrec,et al.  Lightweight probabilistic broadcast , 2003, TOCS.

[17]  Gustavo Alonso,et al.  Processing transactions over optimistic atomic broadcast protocols , 1999, Proceedings. 19th IEEE International Conference on Distributed Computing Systems (Cat. No.99CB37003).

[18]  Mustaque Ahamad,et al.  1/k phase stamping for continuous shared data (extended abstract) , 2000, PODC '00.

[19]  Doug Terry,et al.  Epidemic algorithms for replicated database maintenance , 1988, OPSR.

[20]  Richard A. Golding.  A Weak-Consistency Architecture for Distributed Information Services , 1992, Comput. Syst..

[21]  Kenneth P. Birman,et al.  Bimodal multicast , 1999, TOCS.

[22]  Indranil Gupta,et al.  Fighting fire with fire: using randomized gossip to combat stochastic scalability limits , 2002 .

[23]  Yasushi Saito,et al.  Optimistic replication , 2005, CSUR.