Simon Scarle's giddy enthusiasm (see photo, at right) could be the result of getting to play with Xboxes on the job. More likely though, it's because his recent discovery about the console's graphical processing unit could save thousands of lives.
To do this kind of parallel processing, researchers typically have to book expensive time on a dedicated cluster system or network of PCs; the Warwick team found that the Xbox 360 GPU was able to perform the same functions at a fraction of the time and cost. It's also far more readily available, and comes with the side benefit of being able to play Halo 3 on break.
Not surprisingly, the researcher was once a software engineer at the Warwickshire firm Rare (part of Microsoft Games Studios). Scarle says he first thought up the project while developing a "little shooter game" for Microsoft in which the player guns down enemies in a heart-shaped arena.
This is a highly effective way of carrying out high-end parallel computing on domestic hardware for cardiac simulations. Although major reworking of any previous code framework is required, the Xbox 360 is a very easy platform to develop for and this cost can easily be outweighed by the benefits in gained computational power and speed, as well as the relative ease of visualization of the system.
It turns out that the world's fastest supercomputer, Roadrunner, uses the same processor technology as Sony's Playstation 3. And while Scarle says linking more than one Xbox together is possible, it has yet to be done. "It could be done, but you would have to go over the Internet--through something like Xbox Live--rather than a standard method," he tells the BBC.
Scarle also points to folding@home, a project out of Stanford that devotes the spare processing power of thousands of PCs, Macs, Linux systems, and PlayStation 3s to understanding "protein folding, misfolding, and related diseases." The network has more than 5 petaflops of computing power, aka more than 5,000 trillion calculations per second. (Roadrunner, a $100 million system, can operate at just over 1 petaflop.)
Scarle's discovery does come with one piece of bad news. His research concludes that it is impossible to actually predict certain arrhythmias, because cardiac cell models are affected by a limitation of computational systems called the halting problem.