Thus far, the to increase the industrial might of the U.S. has amounted to little more than an asterisk from a financial standpoint in both the federal budget and the economy as a whole. Market research firm IDC estimates that the public and private sectors spent a combined total of just over $10 billion on HPC systems last year.

To provide some perspective, consider this: since September, the U.S. government has spent $150 billion to keep insurer American International Group Inc. afloat. Meanwhile, HPC resources are inaccessible to many companies that could benefit from using the technology.

On the surface, all looks well with HPC. Federal agencies continue to fund the development of massive systems, such as 's Roadrunner, which last spring became the first machine to break through the . Cray Inc.'s XT5 Jaguar, another system bought and paid for by the government, also topped the petaflop mark this month and was on the latest of the world's fastest supercomputers.

Supercomputers provide something akin to a Second Life for engineers. Instead of building physical models of new products, HPC users can create them in virtual environments and then use the supercomputing horsepower to test how the products work and change design elements and materials as needed. That can cut weeks or even months off of the design and testing process, potentially giving a company a critical edge over its competitors.

But there's a problem in getting HPC capabilities into the hands of companies that create jobs. Thousands of businesses could conceivably make use of the systems, but many can't afford them -- leaving HPC's economic potential largely unrealized.