News
Hadoop on Windows Azure Moves Forward in Microsoft's Embrace of Big Data
- By David Ramel
- March 22, 2013
Microsoft this week released the first public preview of its Hadoop on Windows Azure service, called HDInsight, in a further embrace of Big Data.
After months of private testing, data developers with a Windows Azure account can ask for access to the public preview of the Windows Azure HDInsight Service, which simplifies the deployment and management of Hadoop clusters in Microsoft's cloud.
While using Apache Hadoop on Windows Azure was possible prior to this, the new service removes many of the hoops you previously had to jump through to get things working with the hadooponazure.com beta, as Lynn Langit demonstrated in a July 2012 MSDN Magazine article (full disclosure: I'm technical editor at that magazine).
Microsoft detailed some of the improvements: "This new release isn’t a re-hosting of hadooponazure.com preview offering, but includes a number of enhancements and features ranging from simplified cluster deployment, performance & robustness upgrades, the ability to deploy a cluster of up to 40 nodes and more."
The new service is based on the Hortonworks Data Platform, recently released in beta for Windows Server. Management of Hadoop jobs is provided through interactive JavaScript and Hive consoles. A Hive Open Database Connectivity (ODBC) driver lets developers work with the data via Microsoft Office, and you can use business intelligence (BI) tools such as SQL Server Analysis Services (SSAS), Power View and PowerPivot to analyze the data.
During the preview period Microsoft is discounting the service by 50 percent from
regular pricing. Information on getting started with the new service is available
here, and Shayne Burgess from the HDInsight team provides
more guidance as part of a series of blog posts. Scott Guthrie also wrote about the service on a
post that includes details on other new Windows Azure enhancements for Mobile Services and Web Sites.
What do you think about Microsoft's embrace of Hadoop and Big Data? Comment here or via e-mail.
About the Author
David Ramel is an editor and writer for Converge360.