Advanced platform features and industry-standard models for easy Hadoop cluster management.

KEY FEATURES

Self-Healing Cluster

Run Hadoop in a lights out data center. There are no NameNodes to manage or jobs that need to be restarted. Self-healing protects from multiple failures across different system components including nodes, jobs, and access points.

Consistent Point-in-Time Snapshots

Perform point-in-time recovery of files and tables to overcome user or application created errors. Also configure different snapshot frequencies for different volumes.
Learn More

Disaster Recovery with Mirroring

Set up disaster recovery for all data – both files and tables. Only MapR provides native mirroring to replicate data between production and research environments or between on-premises and cloud infrastructures.
Learn More

Zero Database Administration

Run operational database applications on Hadoop without the need to perform manual administrative tasks such as data splitting and merging or manual compactions.

Advanced Management Console

Complete visibility and protection from the software down to the hardware. Visually monitor node health and also setup alerts and alarms from the MapR Control System (MCS).

Granular Job and Task Diagnostics

Zoom into cluster usage and performance metrics for jobs and tasks with intuitive line charts and histograms. All MCS operations are also available through the MapR CLI and REST APIs.

Volume-Based Data Management

Only MapR supports volumes to make it easy to apply policies across directories and file contents without managing individual files. These policies include data protection, retention, snapshots, and access privileges. Setup quotas for users. Save time with centralized configuration and logging. Also leverage NFS capabilities to access and monitor the cluster.

Multi-Tenancy

Share the cluster. Exploit volumes to create isolated environments for users, groups and applications with different security, hardware or network requirements. Use labels and queues to better manage job execution.

Rolling Upgrades

Eliminate planned downtimes. Perform maintenance or software upgrades on your cluster, a few nodes at a time.