Microsoft unveiled new big data services today that tap the Redmond, Wash., company's globe-spanning Azure cloud. With the Build developer conference in San Francisco as a backdrop, the company took the wraps off new Azure SQL Database elastic databases, Azure SQL Data Warehouse and Azure Data Lake.
Notwithstanding the impending crush of digital information that is expected to be generated by the burgeoning market of Internet of things services and solutions, enterprises are already struggling to capitalize on the data they already have, Corporate Vice President T.K. "Ranga" Rengarajan told eWEEK. "The world is drunk on data," he said.
Today's announcements are part of Microsoft's goal to help enterprises extract "as much value from the data as possible and infuse it with insight," Rengarajan said.
First, Microsoft is working to make Azure more accommodating to cloud independent software vendors (ISVs) that are tasked with managing large numbers of customer databases. "ISVs are saying that they create hundreds and thousands of databases for customers" on Azure, often leading to over-provisioning and productivity penalties that can crimp profits, particularly during peak demand.
The Azure SQL Database elastic databases option enables ISVs to benefit from "predictable business pricing with a pool of unpredictable databases," said Rengarajan. Released today in preview, it allows ISVs and developers to share the resources of an elastic database pool across hundreds or thousands of databases. Also available are developer tools and client libraries that allow developers to create applications that take advantage of the new capability.
Also announced today was Azure SQL Data Warehouse, "the world's first elastic data warehouse," according to Rengarajan.
Based on the massively parallel processing architecture that powers SQL Server, organizations can use the offering to create petabyte-scale data warehouses that dynamically shrink or grow—and can even be paused, revealed Rengarajan—allowing customers to budget according to their query performance requirements. The product earns Microsoft the distinction of being the only company to offer both cloud and on-premise (Analytics Platform System) data warehouses at scale, the company claimed.
Azure SQL Data Warehouse integrates with existing big data tools and services, including the HDInsight Apache Hadoop service, Power BI, Azure Machine Learning and Azure Data Factory. Echoing a common theme at Build, developers won't have to reinvent the wheel to get their big data projects off the ground, Rengarajan said. "You don't have to learn new tricks to leverage this massive power." A public preview is scheduled for June 2015.
Finally, Microsoft kicked off the private preview of Azure Data Lake, the company's "hyperscale repository optimized for big data analytics workloads," said Rengarajan. "There are no size limits on hyperscale files," he added, noting the offering's ability to store petabyte (PB) range files while maintaining immediate read and write access.
Azure Data Lake is aimed at organizations that have based their big data initiatives on the popular open-source Hadoop platform. The Hadoop Distributed File System (HDFS) compatible service is integrated with the company's own Azure HDInsight Hadoop distribution and will be integrated with Revolution-R Enterprise, which Microsoft acquired earlier this year, along with Hortonworks and Cloudera. A public preview will be available later this year.