Extending Mixed Serialisation Graphs to Replicated Environments

Replication has been told to be a solution to provide scalability and high availability in databases. Unfortunately, the cost of ensuring isolated and consistent executions is sometimes too high. Weakest isolation models have proved to be a way to reduce this cost but they can violate some applications transactions isolation needs. In stand-alone systems, models supporting different isolation restrictions for concurrent transaction are used to avoid this dilemma. With this kind of protocols, applications can specify every transaction isolation requirements. However, how to extend these models to replicated systems is still an issue. In this paper we extend Adya's model based on serialization graphs as a first step to construct replication protocols with such a feature.

[1]  Sam Toueg,et al.  Fault-tolerant broadcasts and related problems , 1993 .

[2]  R. Salinas,et al.  SIRC, a multiple isolation level protocol for middleware-based data replication , 2007, 2007 22nd international symposium on computer and information sciences.

[3]  Patrick E. O'Neil,et al.  Generalized isolation level definitions , 2000, Proceedings of 16th International Conference on Data Engineering (Cat. No.00CB37073).

[4]  André Schiper,et al.  Comparison of database replication techniques based on total order broadcast , 2005, IEEE Transactions on Knowledge and Data Engineering.

[5]  Luis Irún-Briz,et al.  Managing Multiple Isolation Levels in Middleware Database Replication Protocols , 2006, ISPA.

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

[7]  Alan Fekete,et al.  Allocating isolation levels to transactions , 2005, PODS '05.

[8]  Josep M. Bernabé-Gisbert Providing Support for Data Replication Protocols with Multiple Isolation Levels , 2007, OTM Workshops.

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

[10]  Luis Irún-Briz,et al.  Managing Transaction Conflicts in Middleware-based Database Replication Architectures , 2006, 2006 25th IEEE Symposium on Reliable Distributed Systems (SRDS'06).

[11]  R. D. Juan-Marín,et al.  Providing Read Committed Isolation Level in Non-Blocking ROWA Database Replication Protocols ⋆ , 2007 .

[12]  Barbara Liskov,et al.  Weak Consistency: A Generalized Theory and Optimistic Implementations for Distributed Transactions , 1999 .

[13]  J. R. Garitagoitia,et al.  Non-blocking ROWA Protocols Implement GSI Using SI Replicas , 2007 .