David Chappell

Opinari

Get the Feed! Subscribe

Introducing LINQ to HPC: Processing Big Data on Windows  
# Monday, May 23, 2011
 
Big data is a big deal. Effectively processing large amounts of unstructured information can have real value, but doing it requires a technology focused on this problem.

If you're interested in the business value of big data, take a look at this recent McKinsey report, which walks through possibilities in several different areas. If you're interested in the technology, you might read the Microsoft-sponsored white paper I've written on LINQ to HPC, a new option for working with big data on Windows.

Part of Windows HPC Server, LINQ to HPC targets problems similar to those addressed by the open source Hadoop. And although it's just been released as a commercial product, LINQ to HPC is in fact a commercialization of Dryad, a technology created by Microsoft Research that's been used within Microsoft for several years. In a world full of big data, it's good to have options for how we make sense out of this sea of bits.


3 comments :: Post a Comment

 


Comments:

Very nice paper. How does the start-up cost and operational costs of HPC compare to Hadoop.
 

I'm not aware of any comparisons that have been done yet, Piotr--LINQ to HPC is brand new.

One obvious observation, though, is that Hadoop will be a natural fit for Linux environments, and so easier for current Linux admins to work with. Similarly, LINQ to HPC will be a natural fit for Windows environments, and so I'd expect it to be cheaper to operate for admins who are already familiar with Windows.
 

Thank you David very much for the white paper on LINQ to HPC and the suggestions on my DryadLINQ CTP paper.

Regards!
Hui
 

Post a Comment


<< Home