Roger's Equations Blog

Roger's Equations

This blog is all about science and technology (with occasional math thrown in for fun). The goal of this blog is to try and pass on the sense of excitement and wonder I feel when I read about these topics. I hope you enjoy the posts.

Previous in Blog: Curl, Divergence, and Gradients.   Next in Blog: The Fiber Optic Future
Close
Close
Close

The Importance of High Performance Computing

Posted May 17, 2005 9:57 PM by Bayes

According to www.top500.org, the top Supercomputer in the world is BlueGene/L, located in Rochester, New York. It consists of roughly 33,000 processors and can perform 70 trillion floating point operations per second (TFlops). It is pedicted by 2010, there will be a supercomputer capable of 1 PFlop, thats 1000 TFlops (Whatis.com). Who knows what kind of computational power we'll be talking about in 2020. Of course, some might say, why care? What could we possibly need such powerful computers for. I think the easiest way to think about it is to think in terms of calculators, since that's what these supercomputers are really, super calculators. I don't believe I've ever calculated a complicated square root by hand, yet I'm in a field where they need to be done all the time. The simple solar powered calculator and later Windows built in calculator has spared me that ordeal. There are hand held calculators that handle integrals and derivatives (as well as websites), though I have to say, none that are easy to use. As a side note, somebody needs to make one that's easy to use, and if it's already out there, market it better, because I'm sick of CRC's tables, but I digress. So with a relatively small amount of computing power, we can do many difficult and time consuming math problems. The problem is, what if you need to calculate millions of integrals through several interations, as is needed when calculating material properties of large molecules or systems with quantum mechanics, or if there are many variables that effect each other as in climate modeling. The point is if we had the availible computing power, we could be calculating and cataloging large systems, noticing patterns, and developing new physics to explain what we see. I think when it comes down to it, we always are gonna need that next generation calculator to solve the next generation of engineering problems.

Reply

Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.

Previous in Blog: Curl, Divergence, and Gradients.   Next in Blog: The Fiber Optic Future
You might be interested in: Math Calculation Software, Thin Clients

Advertisement