A Protocol for Reconciling Recovery and High-Availability in Replicated Databases

We describe a recovery protocol which boosts availability, fault tolerance and performance by enabling failed network nodes to resume an active role immediately after they start recovering. The protocol is designed to work in tandem with middleware-based eager update-everywhere strategies and related group communication systems. The latter provide view synchrony, i.e., knowledge about currently reachable nodes and about the status of messages delivered by faulty and alive nodes. That enables a fast replay of missed updates which defines dynamic database recovery partition. Thus, speeding up the recovery of failed nodes which, together with the rest of the network, may seamlessly continue to process transactions even before their recovery has completed. We specify the protocol in terms of the procedures executed with every message and event of interest and outline a correctness proof.

[1]  Fernando Pedone,et al.  Database replication using generalized snapshot isolation , 2005, 24th IEEE Symposium on Reliable Distributed Systems (SRDS'05).

[2]  JoAnne Holliday.  Replicated database recovery using multicast communication , 2001, Proceedings IEEE International Symposium on Network Computing and Applications. NCA 2001.

[3]  Flaviu Cristian,et al.  Understanding fault-tolerant distributed systems , 1991, CACM.

[4]  Gustavo Alonso,et al.  Using Optimistic Atomic Broadcast in Transaction Processing Systems , 2003, IEEE Trans. Knowl. Data Eng..

[5]  Divyakant Agrawal,et al.  Epidemic Algorithms for Replicated Databases , 2003, IEEE Trans. Knowl. Data Eng..

[6]  Idit Keidar,et al.  Group communication specifications: a comprehensive study , 2001, CSUR.

[7]  Gustavo Alonso,et al.  MIDDLE-R: Consistent database replication at the middleware level , 2005, TOCS.

[8]  Luis Irún-Briz,et al.  Database Replication Protocols , 2005, Encyclopedia of Database Technologies and Applications.

[9]  Francesc D. Muñoz-Escoí,et al.  A Lock Based Algorithm for Concurrency Control and Recovery in a Middleware Replication Software Architecture , 2005, Proceedings of the 38th Annual Hawaii International Conference on System Sciences.

[10]  Bettina Kemme,et al.  Postgres-R(SI): combining replica control with concurrency control based on snapshot isolation , 2005, 21st International Conference on Data Engineering (ICDE'05).

[11]  Ricardo Jiménez-Peris,et al.  Middleware based data replication providing snapshot isolation , 2005, SIGMOD '05.

[12]  Dennis Shasha,et al.  The dangers of replication and a solution , 1996, SIGMOD '96.

[13]  Jim Gray,et al.  A critique of ANSI SQL isolation levels , 1995, SIGMOD '95.

[14]  Francesc D. Muñoz-Escoí,et al.  Design of a MidO2PL database replication protocol in the MADIS middleware architecture , 2006, 20th International Conference on Advanced Information Networking and Applications - Volume 1 (AINA'06).

[15]  Alberto Bartoli,et al.  Online reconfiguration in replicated databases based on group communication , 2001, 2001 International Conference on Dependable Systems and Networks.

[16]  Gustavo Alonso,et al.  A new approach to developing and implementing eager database replication protocols , 2000, TODS.

[17]  Camino de Vera,et al.  A Lock Based Algorithm for Concurrency Control and Recovery in a Middleware Replication Software Architecture , 2005 .

[18]  AlonsoGustavo,et al.  A new approach to developing and implementing eager database replication protocols , 2000 .

[19]  Gustavo Alonso,et al.  Non-intrusive, parallel recovery of replicated data , 2002, 21st IEEE Symposium on Reliable Distributed Systems, 2002. Proceedings..

[20]  Gustavo Alonso,et al.  Understanding replication in databases and distributed systems , 2000, Proceedings 20th IEEE International Conference on Distributed Computing Systems.

[21]  Francesc D. Muñoz-Escoí,et al.  Implementing Database Replication Protocols based on O2PL in a Middleware Architecture , 2006, Databases and Applications.