Open source commercial distributions for Hadoop fall under one of the three categories of innovations shown below. All distributions, as seen in the figure, are essentially packaging the same set of projects from the Apache Hadoop ecosystem. MapR, represented by the third category on the right, includes the latest revisions as well as innovations in a Converged Data Platform that integrates Hadoop and Spark, real-time database capabilities, and global event streaming with big data enterprise storage, for developing and running innovative data applications. Containerized applications can further leverage the MapR Persistent Application Client Containers to securely access and leverage MapR platform services (MapR-FS, MapR-DB, MapR Streams) as a persistent data store.


Enterprise Flexibility

MapR uniquely supports earlier versions of most of the ecosystem projects, allowing enterprises to set their own pace on building their big data capabilities. MapR constantly tests, validates and hardens the open-source projects before including them in its distribution. MapR updates the open source projects in its distribution on a monthly basis, making us the most frequently updated and up-to-date distribution. MapR adds 2-4 new Apache projects per year based on technology production readiness and customer demand. MapR contributions to the open source projects are publicly available through GitHub while MapR binaries can be integrated through Maven repositories.

Hadoop Ecosystem Innovation

MapR also provides key contributions to several projects in the Hadoop stack such as Apache ZooKeeper, Apache Mahout, and Apache Storm. MapR spearheads the development of key open source projects where it can provide unique capabilities to the community. For instance, MapR has incubated and is working with the larger community on the Apache Drill project that has set the ambitious goal of providing millisecond, ANSI SQL query capabilities across databases driven by both centralized schema as well as application-driven dynamic schemas. MapR also initiated open source projects such as Apache Myriad for site-wide resource management, and OJAI for accessing JSON data in Hadoop.
Learn More

Commitment to Open Standards

MapR infrastructure innovations at the file-system layer do not create vendor lock-in for end-users as MapR takes care of ensuring binary level compatibility with Apache projects. It is the same process to switch between any two Hadoop distributions including from and to MapR. More importantly, in-line with its enterprise focus, MapR supports not just Hadoop standard interfaces but also wider industry standards including POSIX, NFS, LDAP, ODBC, REST and Kerberos to name a few.