MapR Distributed File and Object Store (MapR XD)

Extend Converged Data Platform to Create a Cloud-Scale Data Fabric

MapR Distributed File and Object Store (MapR XD) is the Exabyte-scale global data store for diverse data types of files and containers. MapR XD spans multiple edge, on-premises, and cloud environments and uniquely provides in-place analytics for operationalizing insights in real-time.

MapR XD Overview

MapR XD architecture diagram

MapR XD Benefits

  • Global access to the broadest range of data types
  • Widest choice of infrastructure from flash to disks to cloud
  • Automatic Multi-Temperature Distribution Topologies
  • Exabyte scale out design supporting up to trillions of files or objects
  • Self-healing operating model that assumes hardware failures are common
  • Robust management console provides ease of use for administrators
  • Unified management for globally distributed data
  • Multi-tenant security and access control model
  • Data protection and point-in-time recovery with unlimited instant snapshots
  • Creates persistent data store for container applications and analytics
  • Disaster recovery with mirroring to remote sites

Cloud-Scale Data Store

Data growth and variety of workloads are out-pacing the abilities of historical architectures. Changes in application data access requirements, the underlying hardware, software, data types, use cases and cloud all impact what is needed to be the storage and data management platform of the future. MapR XD is an Exabyte scale, global data store built from the ground up to the demanding requirements of today and tomorrow.

One Global Namespace for All Data

The MapR Global Namespace provides a consolidated view into files that are in separate clusters across multiple edge, on-premise and cloud environments enabling development of new location-independent applications that can move across clouds smoothly.

Cloud-Scale Data Fabric for Files and Containers

MapR XD

Multitemperature Data Management

MapR XD provides a single platform for multi-tier, multi-temperature, and multi-cloud data management. MapR Data Topologies enable placing a specified volume’s data in specific nodes or racks in a cluster. Volume topologies can be changed at any time with a simple command regardless of the size or location of the volume. MapR topology movement ensures data is moved from one topology to another (example from hot SSD based nodes to warm HDD based nodes) seamlessly and reliably without impact to any running applications. This enables automated data movement across various temperatures reliably.

Cloud-Grade Resilient Global Data Fabric

MapR Converged Data Platform is built with practically no architectural limitations to achieve scale. MapR can scale to 1000s of hosts and clients across multiple racks. Clusters can connect to each other securely and create a global data fabric across geographically disperse locations including on-premises, at the edge, and multiple clouds. Both metadata and data are distributed and replicated, so no failure, hardware or software, will result in a loss of data or loss of access to data. Instant Snapshots are atomic and consistent, preserving the data at a given point in time. Global mirroring orchestrated offers a robust and efficient way to keep the data well protected.

Integrated Analytics to Operationalize any Data

MapR XD combines analytics and operations into a single platform, enabling intelligent application development. It is the industry’s only infrastructure agnostic platform for intelligent applications that allows enterprises to immediately perform analytics with no movement of data from both legacy sources and new unstructured data sources. MapR XD native support for open APIs (HDFS, POSIX, NFS) ensures availability of files, images, videos, objects, tables, and streams on one platform.

Built for the World of GPUs and NVME

MapR Distributed File and Object Store (MapR XD) is built for the new demanding applications and I/O patterns that are going to be commonplace in the world with modern processing architectures like GPU and modern media architectures like NVME and 3DXPoint. MapR XD is built from the ground up to exploit the extremely fast media types and provide the I/O needed for applications like Real-time IoT operations, next generation intelligent applications, Augmented Reality/Virtual Reality driven research and applications as well as deep analytics.

Download PDF