DataStax’ Brisk, a Hadoop distribution integrated with NoSQL database Cassandra, announced at GigaOM Big Data, is now available for download 46 days after the original announcement -- only one day later than promised.
A Friendlier, Gentler Hadoop
Today, DataStax released Brisk, an open-source Hadoop distribution that adds real-time features and reduces the complexity with deploying and running Hadoop and Hive. Apache Cassandra powers Brisk; the platform offers an integrated environment containing a low-latency database for high-volume web and real-time applications with tightly coupled Hadoop and Hive analytics.
CMSWire spoke with Mike Weir, VP of Marketing for DataStax about Brisk. CMSWire asked Weir what is the unique value proposition for the product. He said,
We are trying to solve the well-known limitations of the Hadoop stack. It’s pretty hard to scale. It has single points of failure. To take a Hadoop deployment from your POC to your production environment is almost a black art.”
Weir went on to say,
Cassandra has some well-known important capabilities in the big data space and combining those capabilities with batch analytic capabilities Hadoop is a really important step in making Hadoop easier to use -- especially the real-time capabilities. In a traditional Hadoop installation, you have to an ETL job. It’s not real-time. You have to stop everything you are doing. With Brisk, the analytics and real-time capture happen simultaneously.”
Brisk uses Hadoop MapReduce and Hive capabilities, but replaces HDFS and HBase with Apache’s Cassandra technology. The Brisk architecture eliminates the HDFS name node and its single-points-of-failure and scalability issues. DataStax’ Brisk also uses Cassandra’s support for multi-datacenter replication, providing the distribution with synchronous or asynchronous replication of data between distributed datacenters via policy definitions.
Wier says Brisk early adopters have shown interest in the real-time capabilities of the platform. DataStax said that 40% of the users to their website have inquired specifically about Cassandra with Hadoop. Werner said, “You can do things actionable things in real-time that impact business.”
When asked about rumors of Yahoo’s entrance into the market, Werner seemed pleased at the news.
Yahoo is the grandfather in this space. We look forward to their entrance in this space. Our work is based off the Yahoo distribution so any work they do will benefit us.”
Organizations considering DataStax’ Brisk should also be aware that DataStax OpsCenter supports the distribution.
Ops center monitoring
OpsCenter provides management and monitoring for Cassandra and the new Brisk Hadoop distribution.
Now that Brisk is officially released, we will be watching to see whether it lives up to the level of buzz that it has created since its original announcement.