DataStax, the commercial leader in Apache Cassandra, today released Brisk, a second-generation open source Hadoop distribution that the company says eliminates the key operational complexities with deploying and running Hadoop and Hive in production. Brisk is powered by Cassandra and offers a single platform containing a low-latency database for high-volume web and real-time applications, while providing tightly coupled Hadoop and Hive analytics.
"The goal in developing and deploying Brisk was to provide customers an easier way to manage large volumes of data, while extracting business insights from that data in the most efficient way possible," says Matt Pfeil, CEO and co-founder, DataStax. "By utilizing Apache Cassandra as the foundation for Hadoop, we've created a distribution where the real-time analysis and creation of data live in the same data-store, effectively eliminating the need to move data or risk systems failing due to work overload."
According to DataStax, Brisk utilizes the best of Hadoop (MapReduce and Hive capabilities), while replacing the weaker pieces (HDFS and HBase) with Cassandra-based technology, resulting in a Hadoop distribution containing a single layer of peer nodes that communicate via a state-of-the-art ‘gossip protocol' for replication and fault-tolerance. This eliminates the HDFS ‘name node' and associated single-points-of-failure and scalability pains.
Brisk also benefits from Cassandra's intrinsic support for multi-datacenter replication. For the first time Hadoop provides automatic synchronous or asynchronous replication of data between two or more distributed datacenters - all controlled through simple policy definitions.
Brisk is compatible with all widely used Hadoop distributions and tools.
For more information go to www.datastax.com/brisk.