MapR Technologies, a big data technology provider and Hadoop specialist offering open source software solutions, announced that Return Path, an e-mail intelligence software company, is using MapR Distribution for Apache Hadoop.
MapR Distribution provides Return Path with enterprise-grade capabilities, reduced administrative costs as well as expert technical support from MapR.
MapR delivers an enterprise-grade platform that supports a broad set of mission-critical and real-time production uses. It brings unprecedented dependability, ease-of-use and world-record speed to Hadoop, NoSQL, database and streaming applications in one unified big data platform.
Return Path had been using Hadoop all this while; however, as their growth accelerated, the limitations of the technology became apparent, said Andy Sautins, CTO of Return Path. After deploying MapR, the company saw 3x performance increase in real-world workloads with MapR.
The compelling features of MapR have reduced the administration overhead, Sautins said. For example, Return Path found that higher density nodes required much less administration on MapR, allowing them to manage nodes of 20 terabytes versus 6 terabytes with their previous solution. This reduces the need of hardware and enables them to run their businesses more efficiently.
Return Path also integrated features like Direct Access (News - Alert) NFS, mirroring for disaster recovery, and volume management into their infrastructure without requiring changes. NFS provides them with the flexibility to use tools other than MapReduce to process data, and simplifies moving data on and off the cluster.
“With MapR, we can handle more drives per node, which translates to better data density in our cluster,” said Shawn Nussbaum, Return Path engineering director. “Prior to using MapR, we ran into limitations with the NameNode due to a large number of files.”
According to Jack Norris, chief marketing officer of MapR Technologies, Return Path’s Hadoop deployment has expanded their business, reduced administrative costs and significantly increased the performance of the system at the company.