Cassandra Adds Hadoop MapReduce
Join the DZone community and get the full member experience.Join For Free
One of the primary new features in Cassandra 0.6 is support for Apache Hadoop. This is a major upgrade for Cassandra, giving it even more "big data" capabilities. The new feature will allow Cassandra to run analytics against its own data using Hadoop's reliable MapReduce framework.
Cassandra 0.6 simplifies its architecture with a new integrated caching row. With the implementation of this new feature, Cassandra no longer needs a separate caching layer. Along with the simplified architecture, Cassandra 0.6 also features a performance boost. The distributed data store can already process thousands of writes per second, and this version's enhancements builds on that number.
"Apache Cassandra 0.6 is 30% faster across the board, building on our already-impressive speed," said Jonathan Ellis, Apache Cassandra Project Management Committee Chair in the press release. "It achieves scale-out without making the kind of design compromises that result in operations teams getting paged at 2 AM." The Storage Team Technical Lead at Twitter, Ryan King, explained Twitter's reasons for using Cassandra: "At Twitter, we're deploying Cassandra to tackle scalability, flexibility and operability issues in a way that's more highly available and cost effective than our current systems."
One of Cassandra's best known features is its lack of any single point of failure. The data store's distributed system smoothly replaces any node that goes down with a new node. The system also has the flexibility to be tuned for more consistency or more availability.
The previous version of Cassandra (0.5) added load balancing and significantly improved bootstrap and concurrency. New tools were also added, including JSON-based data import and export, new JMX metrics, and an improved command line interface. “It's fantastic seeing the Project's community at the ASF grow to match the promise of the technology," said Ellis.
You can download Cassandra 0.6 now on the project's website. For more info on Cassandra, check out "4 Months with Cassandra, a love story."