The growth of Big Data is putting ever more IT organizations in the realm of high-performance computing, even as ever-more-powerful supercomputers are being built at Lawrence Livermore and Oak Ridge National Laboratories, as well as the National Center for Supercomputing Applications at the University of Illinois.
But the lack of scientists and engineers who know how to use supercomputers is throwing a wrench in all this. Dan Lyons, writing at The Daily Beast, quotes Stan Ahalt, director of a supercomputing center at the University of North Carolina-Chapel Hill, saying:
It's not enough to keep building powerful supercomputers unless we have the brains. Think of a supercomputer as a very fast racing engine. We need more drivers to use those engines. ...
[Supercomputers] are far more complex than ordinary laptops and desktop computers. Some use hundreds of thousands of microprocessors. Writing software programs that can split work up among all those microprocessors is a daunting task and requires lots of specialized training, way beyond what the average computer science student gets.
This post at Bio-ITWorld.com outlines some of the potential uses, including personalized genome sequencing, more exact forecasting of severe weather and more precise knowledge of global warming.
There are only a handful of specialized programs in high-performance computing in the country, reports HPCWire.com, and the intricacies of parallel programming - MPI, OpenMP, CUDA, and such - generally are not taught at the undergraduate level.
Wu Feng, a professor of supercomputing at Virginia Polytechnic, says companies have been so desperate for talent that different units of the same company have been courting his graduate students. Most of Feng's students are from other countries and leave the United States for greater opportunities in countries such as India and China, Lyons reports. Indeed, a separate HPCWire post predicts that by November, China will claim more than 100 of the top 500 systems in the world. Five years ago, it had 18 such systems, none in the top 10. Now it has two in the top 10 and 74 systems on the TOP500 list.
Without nuturing the talent, the United States could fall behind on multiple fronts, Lyons reports. The article mention one solution is the Virtual School for Computational Science and Engineering, an online program for graduate students, though the website for that program hasn't been updated in a while.