Titan slays Jaguar in computing comparison, open to R&D comers

In supercomputing, a big player just got bigger, and biotech researchers the world over can benefit. The U.S. Department of Energy's Oak Ridge National Laboratory (ORNL) in Tennessee has unveiled its Titan supercomputer, with some of its incredible computing capacity slated to support life sciences experiments, according to a Bio-IT World report.

Unless you're a supercomputing geek, you're unlikely to get jazzed about the fact that Titan can do 20,000 trillion calculations a second, even though that's about 10 times more than the performance capacity of ORNL's Jaguar system, as CNET reported. And Titan has replaced Jaguar, which was once the world's most powerful computer back in 2009. Yet the big deal for drug researchers and academics is the availability of a portion of Titan's capacity to all qualified comers based on approvals from peers and staff discretion.

If approved, a scientist can tap Titan to run programs that can shine light on whether a drug is likely to hit a certain protein target to treat disease or how well a new formulation supports bioavailability of a therapy. And, on Titan, these programs are likely to run faster than anything available in a scientist's own labs, thanks to technologies such as NVIDIA's K20 GPU processor and technology from Cray. The Titan system is expected to crunch numbers for research of biofuels, astrophysics, climate change as well as life sciences, according to reports from Bio-IT World and CNET

Computing has become more central to scientific research as a larger amount of digital data become available for analysis, prompting the government to bankroll the development of powerful computers such as Titan and the Sequoia supercomputer, which is now ranked as the world's largest. 

"Along with experimentation and theory, we've added computing as a tool in the scientists' toolkit," Sumit Gupta, NVIDIA's general manager of accelerated computing, told Bio-IT World. "Supercomputing is the third pillar of scientific research."

- check out Bio-IT World's article
- and CNET's report