MapR: Powering Hadoop Applications for Real-Time Data

“The creation of Hadoop and the beginning of the big data era traces its roots back to the Google white paper on the MapReduce framework; the MapR name is an acknowledgement of these roots,” says Jack Norris, Chief Marketing Officer, while talking about the rationale behind the company.

Headquartered in San Jose, CA, MapR delivers on the promise of Apache Hadoop with a proven, enterprise-grade platform that supports a broad set of mission-critical and real-time production uses. The MapR Distribution for Hadoop combines over a dozen open source packages such as Apache Mahout, Spark, Hive, Pig and ZooKeeper, with MapR innovations to provide a differentiated Hadoop platform to address enterprise customer applications and uses. It was engineered for the data center with IT operations in mind. MapR enables Hadoop to serve business critical needs for big data applications that cannot afford to lose data and require immediate recovery from node and site failures. It supports these capabilities for the broadest set of Hadoop applications from batch analytics to interactive querying and real-time streaming. “The MapR Data Platform is a modern, true read-write capable, NFS-mountable distributed file-system written in C++ that directly accesses storage hardware—dramatically improving performance and ease of administration. It is also built to process both distributed files and database tables in one unified layer,” says Norris.

To store and process big data, MapR provides a wide range of enterprise-grade capabilities. MapR offers multiple editions of its Hadoop distribution. M3 is the Standard Edition that can be downloaded for free and has unlimited production use. It includes performance advantages, advanced management console, and standard file access. The M5 Enterprise Edition enables full data placement capabilities for increased security, speed and manageability and powers Hadoop applications that meet stringent high availability SLAs even under disaster recovery conditions. The M7 Enterprise Database Edition enables continuous real-time operations and handles high-volume and high-velocity database workloads with batch analytical tasks.