Cloudera Partners with MongoDB to Store Hadoop Data on Their NoSQL DB
Starting from the premise that today “80 percent of enterprise data is unstructured and growing at twice the rate of structured data”, Cloudera and MongoDB have announced a “strategic” partnership meant to provide customers the option to combine Cloudera’s Apache-based Big Data platform with MongoDB’s NoSQL solution.
The first fruit of this collaboration is the certification of the MongoDB Connector for Hadoop for CDH5, enabling Hadoop users to input/output data from/to MongoDB through HDFS. This also “allows MapReduce jobs to run natively within MongoDB," according to Kelly Stirman, Director of Product Marketing at MongoDB, quoted by The Register. Without providing more specifics, Stirman added that this is just the beginning of the partnership and more it is to come. One of the improvements mentioned by the press release is a new version of the MongoDB Connector which is to provide better data management between MongoDB and Hadoop.
Cloudera has seen plenty of news lately, starting with a $160M investment from T. Rowe Price, Google Ventures And Michael Dell and followed by a $740M equity investment by Intel which received an 18% share implying that the company is currently evaluated at about $4B.
They have recently announced Cloudera Live (beta), enabling users to play with Hadoop without having to install anything. Through the Hue interface on top of Hadoop, Cloudera Live lets users create and run Hive, Impala, or Pig queries and jobs over preexisting or user’s data.
Cloudera has certified more than 100 partner products on CDH5, including Splice Machine which provides a RDBMS for Big Data applications running on Hadoop, and Ayasdi which provides real-time query-free Big Data analysis for CDH5.