Showing posts with label windows hpc. Show all posts
Showing posts with label windows hpc. Show all posts

01 June 2010

GNU/Linux *Does* Scale – and How

As everyone knows, GNU/Linux grew up as a project to create a completely free alternative to Unix. Key parts were written by Richard Stallman while living the archetypal hacker's life at and around MIT, and by Linus Torvalds – in his bedroom. Against that background, it's no wonder that one of Microsoft's approaches to attacking GNU/Linux has been to dismiss it on technical grounds: after all, such a rag-bag of code written by long-haired hippies and near-teenagers could hardly be compared with the product of decades of serious, top-down planning by some of best coding professionals money can buy, could it?

On Open Enterprise blog.

15 July 2009

Bill Gates Gets Sharing...Almost

Yesterday I wrote about Microsoft's attempt to persuade scientists to adopt its unloved Windows HPC platform by throwing in a few free (as in beer) programs. Here's another poisoned chalice that's being offered:

In between trying to eradicate polio, tame malaria, and fix the broken U.S. education system, Gates has managed to fulfill a dream of taking some classic physics lectures and making them available free over the Web. The lectures, done in 1964 by noted scientist (and Manhattan Project collaborator) Richard Feynman, take notions such as gravity and explains how they work and the broad implications they have in understanding the ways of the universe.

Gates first saw the series of lectures 20 years ago on vacation and dreamed of being able to make them broadly available. After spending years tracking down the rights--and spending some of his personal fortune--Gates has done just that. Tapping his colleagues in Redmond to create interactive software to accompany the videos, Gates is making the collection available free from the Microsoft Research Web site.

What a kind bloke - spending his *own* personal fortune of uncountable billions, just to make this stuff freely available.

But wait: what do we find when go to that "free" site:

Clicking will install Microsoft Silverlight.

So it seems that this particular free has its own non-free (as in freedom) payload: what a surprise.

That's a disappointment - but hardly unexpected; Microsoft's mantra is that you don't get something for nothing. But elsewhere in the interview with Gates, there's some rather interesting stuff:

Education, particularly if you've got motivated students, the idea of specializing in the brilliant lecture and text being done in a very high-quality way, and shared by everyone, and then the sort of lab and discussion piece that's a different thing that you pick people who are very good at that.

Technology brings more to the lecture availability, in terms of sharing best practices and letting somebody have more resources to do amazing lectures. So, you'd hope that some schools would be open minded to this fitting in, and making them more effective.

What's interesting is that his new-found work in the field of education is bringing him constantly face-to-face with the fact that sharing is actually rather a good thing, and that the more the sharing of knowledge can be facilitated, the more good results.

Of course, he's still trapped by the old Microsoft mindset, and constantly thinking how he can exploit that sharing, in this case by freighting it with all kinds of Microsoft gunk. But at least he's started on the journey, albeit unknowingly.

Follow me @glynmoody on Twitter or identi.ca.

14 July 2009

I Fear Microsoft Geeks Bearing Gifts...

Look, those nice people at Microsoft Research are saving science from its data deluge:

Addressing an audience of prominent academic researchers today at the 10th annual Microsoft Research Faculty Summit, Microsoft External Research Corporate Vice President Tony Hey announced that Microsoft Corp. has developed new software tools with the potential to transform the way much scientific research is done. Project Trident: A Scientific Workflow Workbench allows scientists to easily work with large volumes of data, and the specialized new programs Dryad and DryadLINQ facilitate the use of high-performance computing.

Created as part of the company’s ongoing efforts to advance the state of the art in science and help address world-scale challenges, the new tools are designed to make it easier for scientists to ingest and make sense of data, get answers to questions at a rate not previously possible, and ultimately accelerate the pace of achieving critical breakthrough discoveries. Scientists in data-intensive fields such as oceanography, astronomy, environmental science and medical research can now use these tools to manage, integrate and visualize volumes of information. The tools are available as no-cost downloads to academic researchers and scientists at http://research.microsoft.com/en-us/collaboration/tools.

Aw, shucks, isn't that just *so* kind? Doing all this out of the goodness of their hearts? Or maybe not:

Project Trident was developed by Microsoft Research’s External Research Division specifically to support the scientific community. Project Trident is implemented on top of Microsoft’s Windows Workflow Foundation, using the existing functionality of a commercial workflow engine based on Microsoft SQL Server and Windows HPC Server cluster technologies. DryadLINQ is a combination of the Dryad infrastructure for running parallel systems, developed in the Microsoft Research Silicon Valley lab, and the Language-Integrated Query (LINQ) extensions to the C# programming language.

So basically Project Trident is more Project Trojan Horse - an attempt to get Microsoft HPC Server cluster technologies into the scientific community without anyone noticing. And why might Microsoft be so keen to do that? Maybe something to do with the fact that Windows currently runs just 1% of the top 500 supercomputing sites, while GNU/Linux has over 88% share.

Microsoft's approach here can be summed up as: accept our free dog biscuit, and be lumbered with a dog.

Follow me @glynmoody on Twitter or identi.ca.

17 November 2008

The Super Windows That...Couldn't

One of the more bizarre accusations flung by Microsoft at GNU/Linux over the years is that it doesn't scale. This is part of a larger campaign to portray it as a kind of “toy” operating system – fine for low-end stuff, but nothing you'd want to run your enterprise on....

On Open Enterprise blog.