Bright Computing Highlights Fully Integrated Support for Apache Spark at Spark Summit 2015


Flexible solution allows users to run Spark applications with or without HDFS

By Bright Staff | Jun 15, 2015 3:00:00 PM |



San Jose, California — Bright Computing, the leading provider of vendor-independent cluster and cloud management software, is proud to announce new features of Bright Cluster Manager for Apache Hadoop Version 7.1, which provides fully integrated support for Apache Spark™. Bright Computing will be highlighting the exciting enhancements at the Spark Summit 2015 taking place June 15-17 in San Francisco, CA. The newest version of Bright Cluster Manager for Apache Hadoop bundles, installs, monitors, and manages Apache Spark. Nothing else is needed to deploy, use and maintain the Spark engine for large-scale data processing.

Bright Cluster Manager for Apache Hadoop allows users to run Spark applications with or without use of the Hadoop Distributed File System (HDFS). This means customers can easily leverage their existing file system and rapidly deploy, monitor, and manage production Spark clusters alongside their existing HPC environment.

“Version 7.1 of Bright Cluster Manager for Apache Hadoop is the first enterprise-grade deployment, monitoring, and management solution tailored specifically for Apache Spark,” said Ian Lumb, product marketing manager for Bright. “Our holistic offering uniquely addresses the physical cluster, Hadoop, and Spark in a unified and integrated fashion.”

In making use of HDFS, Spark on Hadoop can be deployed in either standalone or YARN mode. Standalone mode features Spark-specific roles for the Spark master and workers. Role-based management is a key capability for effective operations. In standalone mode, Spark handles its own workloads internally, while Bright Cluster Manager monitors the system and collects metrics. In YARN mode, Hadoop’s YARN handles workloads for Spark. Monitoring is performed using YARN-specific metrics collected by Bright.

Using Spark without Hadoop enables customers to use a variety of HDFS alternatives for big data analytics. This is great news for customers who have already made significant investments in scalable, parallel, distributed file systems, and for whom alternatives to HDFS are critical in their ability to adopt Spark.

Bright Cluster Manager for Apache Hadoop currently supports version 1.3.0 of Apache Spark and support for Version 1.4.0 will follow through routine updates. Bright is committed its customers, providing timely maintenance and support to remain current with new releases.

About Bright Computing

Bright Computing is the leading provider of hardware-agnostic cluster and cloud management software in the world. Bright Cluster Manager® provides a unified solution for the provisioning, scheduling, monitoring, and management of HPC clusters, Hadoop clusters, and OpenStack clouds, and has a presence in more than 500 data centers around the world. Bright Computing’s customer base includes global academic, governmental, financial, healthcare, manufacturing, oil/gas/energy, and pharmaceutical organizations such as Boeing, Intel, NASA, Stanford University, and St. Jude Children’s Research Hospital. Bright partners with Amazon, Cisco, Cray, Dell, Intel, and other industry experts to bring customers the greatest access to leading-edge cluster, server and cloud technology possible.

Pictures and screenshots of Bright Cluster Manager

For more information

Bright Computing, Inc.
Mr. Lionel Gibbons
2880 Zanker Road, Suite 203
San Jose, CA 95134
Tel: +1 408 337 6076