SUNNYVALE, Calif.--(BUSINESS WIRE)--Hortonworks, a leading commercial vendor promoting the innovation, development and support of Apache Hadoop, today announced the availability of an enterprise-proven, high-availability (HA) solution for Hortonworks Data Platform (HDP) 1.0 that empowers operations teams deploying Hadoop on VMware vSphere® with an HA solution that is proven and readily available in the enterprise.
Click to Tweet: @Hortonworks Data Platform Delivers High Availability Solutions for Apache Hadoop @VMware #Hadoop, #bigdata, #HA http://bit.ly/LwaKhS
Hortonworks and VMware have collaborated to provide customers with an enterprise-class solution architecture for high availability. The jointly developed solution pairs Apache Hadoop 1.0, the most proven and reliable distribution to date, and VMware vSphere, the industry-leading virtualization platform for building cloud infrastructures. The solution provides easy-to-use, cost-effective high availability for Hadoop 1.0 clusters that minimizes unplanned downtime and IT service disruption, and eliminates the need for dedicated standby hardware and installation of additional software.
Covering key master services today, this solution is extensible to all master services within a Hadoop cluster. It is the only solution to offer fault tolerance for master services, as well as client and network protocols on the Hadoop 1.0 code line. Users are able to choose whether applications wait for clusters that are healing themselves or not wait and return errors for Hadoop requests.
Key Features:
- Automated Hadoop Distributed File System (HDFS) NameNode and JobTracker failover and restart
- Automated MapReduce failure detection and response to HDFS failover events which allow all running workloads to fail gracefully and restart properly
- Automated detection of server and operating system failures, and restart of virtual machines
- Smart resource management that confirms sufficient resources are available to be able to restart virtual machines on different physical servers in the event of server failure
- Flexible architecture that works with Hadoop 1.0 today and Hadoop 2.0 when it becomes available in the future
“Hortonworks and VMware are delivering a simple, yet robust, enterprise-class HA solution for organizations looking to harden their Hadoop deployments,” said Ari Zilka, chief products officer at Hortonworks. “This new capability furthers our mission of making Apache Hadoop an enterprise viable data platform.”
VMware and Hortonworks have tested and confirmed that the VMware-based Hortonworks Data Platform provides the same throughput as bare metal clusters. The HA solution utilizes the VMware vSphere platform’s HA functionality to monitor and automate failover for NameNode and JobTracker HDP 1.0 services. This enables enterprise data centers to deploy highly available NameNode and JobTracker nodes as virtual machines in their Hortonworks Data Platform 1.0 production cluster.
“As Hadoop implementations for processing big data spread across the enterprise, more and more VMware customers need solutions that offer the security and service level assurance they have come to expect from VMware,” said Jerry Chen, vice president, Cloud and Application Services, VMware. “We are pleased to partner with Hortonworks, a company with domain knowledge of Apache Hadoop, to provide our customers with an enterprise-ready open source solution that enables uninterrupted access to their data.”
Complementary with Hadoop 2.0 HA
Hortonworks is contributing directly to the high-availability solutions being developed in the Hadoop 2.0 code line to deliver a truly stable solution that can be trusted at the enterprise level. This new HA solution for the 1.0 code line complements the HA activity currently in beta on the Hadoop 2.0 code line. Hortonworks is dedicated to keeping this solution 100-percent open source and at no additional cost to the community, outside of existing VMware licenses owned. All code developed as part of this effort is being contributed back to the open source community.
Availability
High availability (HA) solutions for Hortonworks Data Platform will be available with the general availability of the Hortonworks Data Platform. Hortonworks Data Platform 1.0 will be available June 15, 2012, for download at: http://hortonworks.com/products/hortonworksdataplatform/.
For more information, attend our webcast June 26, 2012. Register here http://info.hortonworks.com/HDPLaunchWebinar_Registration.html or contact Hortonworks at http://hortonworks.com/about-us/contact-us/.
Additional Resources
- Find details on the Hortonworks Data Platform
- Keep up with company news and reactions to market trends through the Hortonworks blog
- Get details on Hadoop training and certification through Hortonworks University
- Hear the Hortonworks team discuss Apache Hadoop innovations in the Hortonworks Executive Video Series
- Learn how to become a Hortonworks partner
- Meet the Hortonworks founders and learn the pivotal roles they have played in the Apache Hadoop community
- Understand the transformative power of Apache Hadoop
About Hortonworks
Hortonworks is a leading commercial vendor of Apache Hadoop, the preeminent open source platform for storing, managing and analyzing big data. Our distribution, Hortonworks Data Platform powered by Apache Hadoop, provides an open and stable foundation for enterprises and a growing ecosystem to build and deploy big data solutions. Hortonworks is the trusted source for information on Hadoop, and together with the Apache community, Hortonworks is making Hadoop more robust and easier to install, manage and use. Hortonworks provides unmatched technical support, training and certification programs for enterprises, systems integrators, and technology vendors. For more information, visit www.hortonworks.com.
VMware and VMware vSphere are registered trademarks and/or trademarks of VMware, Inc. in the United States and/or other jurisdictions. The use of the word “partner” or “partnership” does not imply a legal partnership relationship between VMware and any other company.