HADOOP 2: What’s New

Hadoop 2 contains fundamental changes in the architecture that significantly extend the platform, taking the compute platform beyond MapReduce and introducing new application paradigms. Similarly, the storage subsystem has been generalized to support other frameworks besides HDFS. The new version significantly improves scalability and performance in both the compute and storage layers, with disk performance up to five times faster and the compute layer scaling to clusters with more than 100k concurrent tasks. Automatic failover of master servers now provides high availability. We cover all these and other key Hadoop 2 features in this article.