Folding@home: the most powerful and energy efficient supercomputer in the world

Folding@home is a large, world-wide volunteer-based distributed computing project started by my group at Stanford University in 2000. Folding@home has just passed the 4 petaflop scale, making it the most powerful distributed supercomputer in the world. I will talk about how Folding@home works, concentrating on recent advances involving stream computing (running Folding@home code directly on GPU’s and the Cell processor in PlayStation 3’s) and massively parallel computation. In particular, I will also discuss algorithmic advances that we have made to map problems to this architecture. I will conclude with a discussion of the future of molecular simulation and a growing paradigm shift from anecdotal simulations, to full-blown statistical models with much greater power and efficiency than traditional computational methods.

Presenter(s)

Prof. Pande's current research centers on the development and application of novel grid computing simulation techniques to address problems in chemical biology. In particular, he has pioneered novel distributed computing methodology to break fundamental barriers in the simulation of kinetics and thermodynamics of proteins and nucleic acids. As director of the Folding@Home project (http://folding.stanford.edu), Prof. Pande has, for the first time, directly simulated protein folding dynamics with quantitative comparisons with experiment, often considered a "holy grail" of computational biology. His current research also includes novel simulation methods for high accuracy drug binding affinity calculations, protein misfolding, and related diseases such as Alzheimer' s Disease.

Prof. Pande received a BA in Physics from Princeton University in 1992. There, he was first introduced to biophysical questions, especially in his undergraduate thesis research with Prof. Philip Anderson, a Nobel Laureate in physics. In 1995, he received his PhD? in physics from MIT, studying as a NSF Fellow under Profs. Toyoichi Tanaka and Alexander Grosberg. At MIT, Prof. Pande's research centered on statistical mechanical models of protein folding and design, suggesting novel ways to design protein sequences to have the desired stability and folding properties. As a Miller Fellow working with Prof. Daniel Rokhsar at UC Berkeley, Prof. Pande extended this methodology to examine atomistic protein models, laying the foundations for his later work at Stanford University.

Prof. Pande has won numerous awards, including the Irving Sigal Young Investigator Award from the Protein Society (2006), the MIT Indus Global Technovator's Award (2004), a Henry and Camile Dreyfus Teacher-Scholar award (2003), being named to MIT's TR100 (2002), and named a Frederick E. Terman Fellow (2002). Prof. Pande is currently an Associate Professor of Chemistry and (by courtesy) of Structural Biology at Stanford University.