C-JDBC: Flexible Database Clustering Middleware

Large web or e-commerce sites are frequently hosted on clusters. Successful open-source tools exist for clustering the front tiers of such sites (web servers and application servers). No comparable success has been achieved for scaling the backend databases. An expensive SMP machine is required if the database tier becomes the bottleneck. The few tools that exist for clustering databases are often database-specific and/or proprietary.

Clustered JDBC (C-JDBC) addresses this problem. It is a freely available, open-source, flexible and efficient middleware for database clustering. C-JDBC presents a single virtual database to the application through the JDBC interface. It does not require any modification to JDBC-based applications. It furthermore works with any database engine that provides a JDBC driver, without modification to the database engine. The flexible architecture of C-JDBC supports large and complex database cluster architectures offering various performance, fault tolerance and availability tradeoffs.

We present the design and the implementation of C-JDBC, as well as some uses of the system in various scenarios. Finally, performance measurements using a clustered implementation of the TPC-W benchmark show the efficiency and scalability of C-JDBC.