InterSystems has announced CACHE 2010, a new release of its CACHE object database. CACHE 2010 targets Java-based high-performance processing combined with persistent storage for event processing systems.
Among other features, CACHE 2010 introduces Database Mirroring which provides automatic failover between any two CACHE-based systems, without the need for specialized (and expensive) storage and networking hardware and software. Key benefits include:
- Flexibility for Planned Downtimes. Configuration changes or operating system upgrades affecting one CACHE-based application, for example, can be executed with minimal impact on overall application availability or performance against Service Level Agreements.
- Minimized Risk. In contrast to traditional replication solutions with the inherent complexity of their configuration requirements, CACHE Database Mirroring utilizes a simple, easy-to-implement model. This approach removes complexity from the configuration equation. And, by using logical data replication, mirroring reduces risks such as out-of-order updates and carry-forward corruption that are possible in the physical replication technologies used by other systems.
- Business Continuity Support. Mirrored databases can be housed in separate data centers, so continuity of key business operations is ensured in the event of a disaster.
CACHÉ 2010 is available for Windows, Linux, Mac, UNIX and OpenVMS platforms.
The update also includes support for Java with the release of CACHE eXTreme for Java which provides direct Java access to the CACHE multidimensional database engine. InterSystems claims that CACHE eXTreme for Java delivers performance equal to an in-memory database while also delivering the persistence of both historical and transactional data that is required for CEP and event-driven SOA.
CACHE eXTreme makes it possible for Java developers to access data efficiently as multidimensional data structures to obtain the highest possible application performance. With sparse storage techniques, sophisticated cache management, and high concurrency rates, CACHE-based applications can be scaled to many thousands of clients without sacrificing high performance, says InterSystems. Another alternative is eXTreme event persistence where developers store Java objects in the CACHE database. This approach automatically creates a Java binding and data can be accessed through Java using objects or SQL for maximum development flexibility.