MapR Extends Lead in Operational Hadoop Capabilities

MapR Technologies Inc.announced the availability of the 4.0.1 version of the MapR distribution including Hadoop.

Providing open source support and enterprise capabilities, the MapR release expands real-time capabilities for use cases including operational applications, interactive queries and stream processing with new versions of Apache Drill, Apache Spark, and Apache HBase, all while leveraging Hadoop 2.4, including YARN.

“Hadoop is generating increasing interest for use in business-critical applications that rely upon low latency and consistency,” said Donnie Berkholz analyst, RedMonk. “This latest release from MapR targets these needs with improved support for real-time processing through the incorporation of new tools as well as new releases of existing ones.“

“Organizations can now maximize business impact and minimize risk by running operational applications along with real-time analytics on Hadoop,” said Tomer Shiran, VP product management, MapR. “MapR continues to lead the market by integrating the latest open source components into a mission critical, multi-tenant platform with self-healing HA, DR and data protection capabilities.”

Unbiased Open Source Approach

MapR provides users with flexibility to pick the right tool for the job with an unbiased set of open source software on Hadoop. The latest MapR Distribution including Hadoop now contains:
Multiple batch processing frameworks including MapReduce 1.x and 2.x (YARN-based), and Spark (0.9, 1.0.2).
Five SQL-on-Hadoop technologies: Hive (0.11, 0.12, 0.13), Drill (0.5), SparkSQL (1.0.2), Impala (1.3.1) and certified integration with HP-Vertica
Two NoSQL technologies: HBase (0.94.21, 0.98.4), MapR-DB
Three machine-learning and graph libraries: Mahout (0.8, 0.9), MLLib (0.9, 1.0.2), GraphX

The MapR Difference

A single MapR cluster can run thousands of disparate jobs a day and accommodate users and departments with different needs. MapR offers several operational features including:
Backward compatibility across versions of open source software packages and ability to upgrade at an organization’s own pace
Heterogeneous processing of MapReduce 1.x and 2.x (YARN-based) applications running on the same set of nodes
Advanced multi-tenancy capability to isolate and protect departmental data on specific nodes and the flexibility to schedule jobs only to those nodes via the label-based scheduling feature, which is now available for YARN jobs as well
Fine-grained resource management to enable consistent high-performance applications supported by unique resource management of disk I/O along with memory and CPU, when deployed via YARN
Comprehensive security through its easily deployable end-to-end, wire-level security that now includes all YARN applications as well
A no-NameNode architecture which enables self-healing clusters that can handle multiple simultaneous failures as well as an unlimited number of files and tables
Snapshots that provide point-in-time consistency enabling system-of-record data to be stored on Hadoop
BC with immediate remote-site recovery for batch and real-time applications
NFS support that enables true random read-write capability so applications can write to and update data on Hadoop even while the analysis is in progress

Version 4.0.1 of the MapR distribution is available.