IBM and Hortonworks today announced the availability of Hortonworks Data Platform for IBM's Elastic Storage Server and IBM Spectrum Scale. The collaboration aims to enable IBM clients to use existing and future investments in IBM storage to deploy Hadoop-based Big Data applications.
Bringing Hortonworks Data Platform to Elastic Storage Server and IBM Spectrum Scale has three key benefits: better storage efficiency, hybrid storage, and high performance. In terms of the first benefit, Elastic Storage Server uses erasure coding that eliminates the need to have multiple data copies. Second, the combined service extends on-premises storage to the cloud, delivering economic, security, and accessibility benefits. Third, Elastic Storage Server delivers high-speed data throughput.
"Because Spectrum Scale is a unified file object and Hadoop Distributed File System environment, you can put things directly into it as objects, and they become immediately available for the Hadoop instances as well," says Doug O’Flaherty, portfolio marketing manager at IBM storage and software-defined infrastructure. "One of the things that is tremendous in terms of our efficiency and performance is that the data ingest, the data that you want to have analyzed, can all start off in Spectrum Scale, and the results can all remain in the Spectrum Scale environment. The traditional problem with Hadoop adoption has often been that it has to be its own analytics silo and data has to go in, be worked on, and then come out."
In September of last year, Hortonworks and IBM announced Hortonworks Data Platform for IBM Power Systems. That effort combined Hadoop and Spark distribution with Power Systems performance, with the goal of empowering clients to quickly derive business insights from their structured and unstructured data.
"We've been having a whole initiative around data and analytics all across IBM, and the Power team has been leading the way through the open data platform initiative, which is about giving our clients the most amount of choice in the open source and commercially supported offerings that they use for analytics. It's part of a broad approach that IBM has—it's not going to be just one kind of application that you use, you're going to be looking at data in many different ways," O'Flaherty says. "We're excited about this storage announcement because our customers who are already running Spectrum Scale who really have an affinity towards Hortonworks will now have the ability to take that data that they're using for other types of analytics, simulations, or data structures and now be able to do Hortonworks…on top of that."