One of the more noteworthy things about Hadoop is how quickly the platform is evolving in terms of its ability to manage data. For example, one of the more sophisticated attributes of version 2.0 of Hadoop is that it provides resource management that is isolated from the underlying processing components. That capability is otherwise known as Yet Another Resource Manager (YARN), and in Hadoop, it opens up a lot of interesting ways to more efficiently integrate data.
Among the first to take advantage of YARN as a mechanism for more efficiently integrating Hadoop data is SnapLogic, which today announced that its integration platform as a service iPaaS environment can now run directly on top of YARN.
Maneesh Joshi, senior director of product marketing for SnapLogic, says SnapReduce 2.0 allows citizen integrators to take advantage of a set of visual tools to integrate as many as 160 data sources out of the box using Snap connectors.
SnapReduce 2.0 also brings the SnapLogic Hadoop Distributed File System (HDFS) in line with updates made to HDFS in Hadoop 2.0, while making it easier to integrate Hadoop data and analytics results with other applications and data stores, such as statistical software, business intelligence applications or data visualization tools.
While SnapReduce 2.0 can be deployed on any instance of Hadoop 2.0, Joshi says it’s becoming apparent that data warehouse services such as Amazon Redshift are fundamentally changing the economics of storing data. As such, Joshi says it’s only a matter of time before most data warehouse applications move into the cloud.
In fact, a recent survey of more than 100 U.S. companies with more than $500 million in revenue that was conducted by TechValidate Research on behalf of SnapLogic found that 56 percent of these companies already use four or more software as a service (SaaS) applications. The survey also found that 22 percent said they expect 25 to 50 percent of all their business apps will be “delivered as a cloud service” by 2015. As the center of data gravity essentially moves to the cloud, Joshi says it’s only a matter of time before customers deploy SnapLogic in the cloud.
Of course, it’s still not clear to what degree Hadoop will either become an extension to the data warehouse or completely replace it. But there’s no doubt that Hadoop will play a significant role in the enterprise, which will require some changes to how IT organizations approach data integration both now and into the future.