Blog Post

Hortonworks and Microsoft bring open-source Hadoop to Windows

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

There’s probably no better way to open up big data to the masses than making it accessible and manipulatable — if that’s a word — via Microsoft Excel. And that ability gets closer to reality Monday with the beta release of Hortonworks Data Platform for Windows. The product of a year-old collaboration between Hortonworks and Microsoft is now downloadable.  General availability will come later in the second quarter, said Shawn Connolly, Hortonworks’ VP of corporate strategy,  in an interview.


The combination should  make it easier to integrate data from SQL Server and Hadoop and to funnel all that into Excel for charting and pivoting and all the tasks Excel is good at, Connolly added.

He stressed that this means the very same Apache Hadoop distribution will run on Linux and Windows. An analogous Hortonworks Data Platform for Windows Azure is still in the works.

Microsoft opted to work with Hortonworks rather than to continue its own “Dryad” project, as GigaOM’s Derrick Harris reported a year ago. Those with long memories will recall this isn’t the first time that Microsoft relied on outside expertise for database work. The guts of early SQL Server came to the company via Sybase.

The intersection of structured SQL and  unstructured Hadoop universes is indeed a hotspot, as Derrick Harris reported last week, with companies including Hadoop rivals Cloudera and EMC Greenplum all working that fertile terrain. That means Hortonworks/Microsoft face stiff competition. This topic, along with real-time data tracking, will be discussed at GigaOM’s Structure Data conference in New York on March 20-21.

3 Responses to “Hortonworks and Microsoft bring open-source Hadoop to Windows”

  1. dlmaniac

    Hadoop is overrated.

    What do you call a big pile of data sitting in a Hadoop storage that cannot be analysed effectively due to its non-structural nature? It’s called a big pile of JUNK w/ all due respect, and that sadly is the state of Hadoop.

    Hadoop’s infrastructure is tiltled toward map & reduce algorithm rather than analytical work. It certainly can store a lot of data but just does not do much useful things to it, which is not surprising as there’s no short cut here when it comes to data analysis: If you don’t organize your data up front then prepare to pay the price down the road when you try to analyse it. GUARANTEED.

    Haddop feels like a typical Silicon-Valley hype (e.g. Facebook, Zynga and so on) rather than some solid products ready to kick butt. Sorry, no sale here.