Microsoft: Big data, meet big compute

Microsoft: Big data, meet big compute

Summary: Microsoft's high-performance computing team is now the 'Big Compute' team. Expect more Windows Azure deliverables from them in the future.


There hasn't been much news from Microsoft's High Performance Computing (HPC) team since December 2012, when the company delivered HPC Pack 2012.


But on July 29, via a post to the "Windows HPC Team Blog," the HPC team noted that it's now going by the name "Big Compute." The Big Compute team will be part of the new Enterprise and Cloud engineering group in the re-org'd Microsoft, going forward. The team will continue to work closely with Microsoft Research and academic/labs partners outside the company, the post added.

"Big Compute applications typically require large amounts of compute power for hours or days at a time. Some of our customers describe what they are doing as HPC, but others call it risk analysis, rendering, transcoding, or digital design and manufacturing," explained post author Alex Sutton, Group Program Manager of Big Compute.

Sutton said the Big Compute team will continue to work on the HPC Pack for Windows Server clusters, as well as "new Big Compute scenarios in the cloud" with Windows Azure. There will continue to be new features and new releases from the team "on a regular basis," he said.

Microsoft's HPC software allows users to run HPC applications on HPC clusters that include on-premises compute nodes, part-time servers and resources running on Windows Azure. Last year, as part of its move to simplify and reduce the number of Windows Server SKUs, Microsoft discontinued its Windows Server HPC family, and opted to provide the HPC functionality as a supplementary "pack" for Windows Server.

"Windows Azure is evolving to provide a range of capabilities for Big Compute and Big Data," Sutton said.

The HPC/Big Compute/supercomputing space remains a battleground for Windows and Linux. And right now, Microsoft looks to be the underdog in this fight

Topics: Big Data, Cloud, Linux, Windows, Windows Server


Mary Jo has covered the tech industry for 30 years for a variety of publications and Web sites, and is a frequent guest on radio, TV and podcasts, speaking about all things Microsoft-related. She is the author of Microsoft 2.0: How Microsoft plans to stay relevant in the post-Gates era (John Wiley & Sons, 2008).

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • underdog is an understatement.

    Roadkill would be more accurate.

    Windows is just too inflexible, and too much in the way for high performance computing. Both are why UNIX lost that market.
  • Microsoft Needs More Drive Letters

    I wonder how they access all that Big Data with just 26 drive letters.
    • OK, you know this is on a cloud service, right?

      So any issue with drive lettering is theirs not yours (leaving aside the fact that mounting storage units via UNC paths has been in Windows longer than Loverock's "Telnet on linux" problem has been fixed.
  • Just Stop, its hurting my sides to laugh so hard

    Big data, big compute. Microsoft in the same sentence? Back in the day, one novell server or one unix server running all the apps, and sharing all the files. Put in windows servers and it took 6 servers to do the job of one novell or unix box? Oh yeah, I forgot. Microsoft is better, right?
  • This section should be renamed from "Comments" to "Flame Wars."

    Too bad... I used to learn stuff from comments.
  • The paradox of Windows and High Performance

    The words "High performance" in association with Microsoft are contradictory to an extreme in regard Microsoft's inability to gain any traction in the "real" HPC markets of Super computing - Top 500, Financial Stock Exchanges or even very high end dominant Software-as-a-Service or Cloud infrastructure, as only a few notations, when compared closely with the fire brand results and status of UNIX and Linux systems in that computing space.

    Just recently, as an example, Netflix chose FreeBSD based Network gateway appliances for streaming millions of movies per minute to customers globally, when Windows HPC could not even get off the ground in trials.

    I guess the references are for Microsoft technology only clients who experience 'higher" performance than with their early 2003-2008 Windows implementations, and therefore have no knowledge of the *NIX world passing them by.