By Matthew Dublin

We never thought we would see the day when Dell would one-up Cray when it comes to large-scale scientific computing, but according to Axel Kohlmeyer, director of the Institute for Computational Molecular Science at Temple University, the proof is undeniable. Kohlmeyer and his team are using two Dell clusters at the National Center for Supercomputing Applications (NCSA) to model surfactants, molecules commonly used in household cleaning products but which also have the potential to control drug delivery in the body. The Temple group has been using two Dell clusters, named Abe and Lincoln, to test a relatively new open-source molecular dynamics simulation software called HOOMD-blue, written expressly for use on GPUs.

"The outcome is quite spectacular ... With two GPUs we can run a single simulation as fast as on 128 CPUs of a Cray XT3," says Kohlmeyer, who is also using Abe to simulate self-assembling dendrimeric molecules which they hope can one day be used to build customized "containers" for drugs. After the simulations have been thoroughly tweaked, they're going to move over to the more powerful Lincoln cluster, which is outfitted with NVIDIA Tesla S1070 Accelerator Units. "We will be able to screen potential modifications to the individual molecules on the computer and save people tons of hours and money in the lab," he says. "A machine like Lincoln is perfect for that, as one would need to run many variations at the same time...We can try things that were undoable before. It still blows my mind."

Here's a video of the Temple team discussing their work: