News

Submissions sought for symposium on chemical computations on GP-GPUs

CyberChem partners Thom H. Dunning, Jr. and Volodymyr Kindratenko, National Center for Supercomputing Applications, University of Illinois at Urbana-Champaign; Todd J. Martinez, Stanford University; Robert J. Harrison, University of Tennessee and Oak Ridge National Laboratory; and Jeffrey Madura, Duquesne University are organizing a symposium on "Chemical Computations on General Purpose Graphics Processing Units" to be held in conjunction with the 240th ACS National Meeting & Exposition in Boston (Aug. 22-26). The symposium will provide technical presentations from the companies that are advancing the development of general purpose graphics processing units (GP-GPUs), discussions of the challenges involved in effectively programming GP-GPUs, and presentations on the use of GP-GPUs in a broad range of chemical applications.

Although exact solution of the equations of quantum and classical mechanics is not possible, substantial progress has been made over the past two decades in the solution of these equations using advanced computing technologies. It is now possible to compute many properties of small molecular systems with an accuracy that rivals that obtained in laboratory measurements, and, for larger molecular systems, computations can provide insights into the details of molecular processes that cannot be obtained from experiment alone. As computing technologies continue to advance, significant progress is possible toward many long sought goals in chemistry, e.g., from a deep understanding of the factors that determine the structure, energetics, and dynamics of molecules to the rational design of chemicals, materials, and drugs.

But, computing technologies are changing--processors based on a single core are being left behind, replaced by multi-core and many-core processors. The impact of these technological changes must be understood in order to take full advantage of the opportunities that they offer. One of the most remarkable of these advances has been the development of the General Purpose Graphics Processing Unit (GP-GPU), which uses hundreds of simplified compute cores to achieve speeds well beyond that available with conventional CPUs. A number of chemistry computations have now been implemented on GP-GPUs, achieving speed-ups of one to two orders of magnitude.

Invited presentations include Michael Klein (Temple University), Mark Gordon (Iowa State University), Alan Aspuru-Guzik (Harvard University), Vijay Pande (Stanford University), Alexey Onufriev (Virginia Tech), Alex Travesset (Iowa State University), Millard Alexander (University of Maryland), Ivan Ufimtsev (Stanford University), Wen-mei Hwu (University of Illinois), John Stone (University of Illinois), and Scott Le Grand (NVIDIA).

Researchers interested in presenting at the symposium should submit an abstract per instructions posted on the ACS website by April 5. All submissions will be reviewed by the symposium organizers, and the accepted submissions will be presented as either oral presentations or posters.

Questions? Contact Volodymyr Kindratenko at kindr@ncsa.illinois.edu.