Why is this page text-only?


App-Rising.com covers the development and adoption of broadband applications, the deployment of and need for broadband networks, and the demands placed on policy to adapt to the revolutionary opportunities made possible by the Internet.

App-Rising.com is written by Geoff Daily, a DC-based technology journalist, broadband activist, marketing consultant, and Internet entrepreneur.

App-Rising.com is supported in part by AT&T;, however all views and opinions expressed herein are solely my own.

« The Impossible Dream of Competitive Broadband Marketplaces In Unserved Areas | Main | Bring on the Exaflood! »

October 11, 2007 8:07 AM

Research Gets A Heavy Dose of Bandwidth

A couple of big announcements last week from the world of Internet-enabled research.

On Monday, Google and IBM teamed up to announce the launch of a program that will open up high powered computing clusters to universities.

These computing clusters are sometimes referred to as clouds and/or grid computing. I wrote a post about this a few weeks back. The gist of cloud/grid computing is by using the Internet you can stitch together the processing power of multiple computers to process large data sets more quickly than any single computer could.

In the case of this Google/IBM program, 1600 computers are being made available across three locations.

The reason for making this computing power available is twofold: one, to give students an opportunity to learn firsthand how to program software that works in a grid computing environment; and two, to give researchers access to a computer grid that can help crunch all sorts of numbers.

For example, in Maryland, the cloud will be used “to create a system for automatically translating text in difficult foreign languages such as Chinese and Arabic,” according to this WashingtonPost.com article.

Networks that link physically distant computers and combines their processing power and connectivity are fascinating to me because perhaps nowhere else can we find a more compelling reasons for broadband connectivity: an opportunity to create the world’s most powerful virtual supercomputers and in doing so made possible incredible new tools in the drive to solve the world’s most pressing problems.

The other major news was Internet2’s announcement of their recently upgraded network.

Internet2 is a high speed network that links more than 200 college campuses with ultra-highspeed connections for the purposes of enabling cutting edge research on a host of topics. Some private businesses have access to and use the network as well, but it’s a largely academic focused endeavor.

Part of their announcement was the fact that the Internet2 is now operating at 100Gbps, with the possibility of increasing that speed tenfold in the near future.

But perhaps even more notable is the ability for Internet2 members to now provision their own dedicated 10Gbps connections for limited periods of time on demand. That means if you’re on Internet2, you can now dial up a connection with another university and push a terabyte of data in less than 15 minutes.

Now that might seem just a bit excessive to the average home user, for whom the term “terabyte” is probably an utterly foreign and unfathomable concept.

But research nowadays generates a ton of data. As analytic technologies improve and computers get faster, the amount of bits and bytes that researchers are analyzing in order to more fully understand how the world works is increasing exponentially.

What I’m not sure about at this time, though, is whether these on-demand 10Gbps connections will be used for anything more than moving large datasets quickly. I’m curious to find out if there are any applications being run that require that much bandwidth to push data in real-time.

I may not know the answer now, but never fear! I’m on the hunt to find out.


TrackBack URL for this entry:

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)