As a part of their any data, any size, anywhere initiative, at this year’s PASS 2011 conference in Seattle Ted Kummert, Microsoft Corporate Vice President Business Platform Division announced Microsoft’s intension to deliver Hadoop based data management to Windows Server and Windows Azure. Hadoop is a distributed computing platform that is written in Java. It uses a MapReduce technology to distribute computing resources across clusters of computers to rapidly process extremely high volumes of data.
At first Microsoft’s embracing of Hadoop may seem a bit confusing but when you view like their incorporation of OLAP services back in SQL Server 7 you can see that support for Hadoop is really another step along the evolutionary path of SQL Server becoming a complete enterprise data platform. Hadoop will enable the processing of high volumes of data as well as unstructured data. A CTP is planned to be released before the end of the year.
The Hadoop implementation will be integrated with both SQL Server relational databases as well as Analysis services. Microsoft announced Hadoop Connectors for SQL Server and the Parallel Data Warehouse. Customers can use these connectors for bi-directional integration of Hadoop with SQL Server’s relational and BI data. Hadoop data will be able to be consumed using Microsoft’s BI tools, including Microsoft PowerPivot and Power View.