Research by Lee-Ping Wang, postdoctoral researcher, Stanford University
Water is the universal medium of life. But when it comes to understanding its behavior on the molecular level, many mysteries remain. The leading model of water, used across many fields of science, is 30 years old and is known to be inaccurate for many research problems.
Researchers from Stanford and The University of Texas at Austin used Stampede to develop and test an improved algorithm for representing water. A better representation of water will assist in finding new drugs and understanding how cells behave. It will also help answer fundamental questions like why water is densest at 4 degrees Celsius and why skaters can slide smoothly across the surface of ice.
Texas Unleashes Stampede
The researchers used Stampede’s graphics processing units (GPUs) as computing engines, simulating the characteristics of water in a variety of temperatures and conditions. The GPUs allowed them to build statistical models that better represented water for diverse applications, from drug design to atomic chemistry. Stampede includes several processing technologies that allow researchers to explore problems in diverse ways, from molecular simulations to high-resolution visualizations to the mining of massive datasets.
“It’s the combination of speed and volume that makes Stampede such a powerful system. Without it, we wouldn’t be able to develop and test our new water model and show how effective it can be,” says Lee-Ping Wang, a postdoctoral researcher from Stanford University.
