Apache Hadoop, the preeminent open source platform for storing, managing and analyzing big data, is a rapidly growing ecosystem around which supporters, like Hortonworks, are building enterprise solutions.
Hortonworks is a leading commercial vendor of Apache Hadoop that just unveiled the newest version of the Hortonworks Data Platform (HDP), claiming to be the industry’s only 100 percent open source platform powered by Apache Hadoop. HDP 1.1 strengthens Apache Hadoop as a reliable, next-generation enterprise data platform for deriving greater business value and insight from big data, which includes Red Hat Enterprise Linux and the High Availability Add-On, Flume integration, monitoring API enhancements and core platform performance improvements.
HDP 1.1 is built on Apache Hadoop 1.0, the most stable and reliable Hadoop code available today that provides enterprise organizations with a proven and trusted data management infrastructure for harnessing the power of big data.
HPD 1.1 features a number of enhancements designed to improve Apache Hadoop deployments in enterprise environments, such as high availability with Red Hat Enterprise Linux and the high availability add-on that improves reliability and stability while broadening the enterprise readiness of Apache Hadoop; Flume Integration which enables expanded streaming data capture for analysis within the Hortonworks Data Platform; Improved Performance for faster read and write in HDFS speeds data capture and delivery within the platform; Improved MapReduce execution performance means that jobs process data more quickly; Open Management Console for Monitoring API Enhancements which delivers easier and deeper integration into third-party management tools and systems to gain insight into performance and assure availability, provision nodes in a cluster and perform ongoing maintenance of the Hadoop platform.
“With multiple high availability options and Flume integration, Hortonworks is delivering a truly non-proprietary 100-percent open source version of Apache Hadoop that is extremely stable and enterprise-ready,” said Ari Zilka, chief products officer, Hortonworks.
“Coming on the heels of the HDP 1.0 introduction this summer, HDP 1.1 continues our development of an ecosystem-friendly Apache Hadoop platform that is more easily integrated with our partners’ applications and our customers’ existing enterprise systems. By making Apache Hadoop easier to consume and use in enterprise environments, Hortonworks is empowering organizations to derive more business value from their big data.”
In other Hortonworks news, the Yahoo spin-off recently published a series of blog posts that explains in detail what Hadoop YARN, a system for managing distributed applications, is all about. YARN, which has been dubbed as the Next-Generation MapReduce, could help Hortonworks reach a larger segment of the big data market, including those reluctant to try Hadoop in the past.
Speaking of MapReduce, a model for processing large data sets using distributed computing and clusters of servers that was created by Facebook in conjunction with Yahoo, Facebook is working on the Hadoop-leveraged “Project Prism,” which would allow them to have their servers in various locations but they will still be able to view their data in a single location. The idea for the project came when it became clear that their data centers aren’t going to be able to hold all the data and servers in just one location.
Apache is another big proponent of Hadoop, recently revealing Apache Drill. It will allow business people to get answers to specific questions from the Apache Hadoop Big Data databases in seconds. But like Project Prism, Apache Drill is far from being ready for the enterprise level.