Computing That Serves

Bargteil: Physics Based Animation; Hall: Autotuning in the Multi-core Era


Thursday, December 11, 2008 - 11:00am


Adam W. Bargteil
Assistant Professor, Computer Science, U. of Utah
Mary Hall
Associate Professor, Computer Science, U. of Utah

Bargteil: Physical simulation has been an important part of computational science for fifty years.  It has allowed engineers to determine whether a bridge would collapse or an airplane would fly, facilitating new designs, saving many dollars and potentially many lives.  Over the last decade physical simulation has proved to be an important tool in computer animation, generating a wide variety of extremely realistic motion of complicated, high degree-of-freedom systems.  The focus on visual detail and fidelity in computer animation presents challenges and opportunities uniques to the field.  In this talk I will give an overview of my work in this area with an emphasis on the visual results (i.e. lots of cool videos).

Hall: We are entering an era where every computer, whether embedded,  laptop, desktop, server or supercomputer, is a parallel computer. Technology trends and concerns about both design complexity and power have pushed computer architecture in this direction, but now there is an enormous burden on software developers to exploit this parallelism or risk slowing down yesterday's applications on tomorrow's hardware. This talk will describe autotuning, a new approach to software technology to ease the programming burden in the multi-core era. Autotuners experiment with a set of alternative application mapping strategies to select the mapping that best exploits architectural features such as deep memory hierarchies, specialized compute engines and multiple cores. Such programming tools increase programmer productivity by reducing the effort of porting to new architectures, and empowering the programmer to maintain code that is simpler, and architecture independent.  Our research involves developing compiler-based autotuning technology, and applying it to application domains that include molecular dynamics, biomedical imaging, signal processing and social networks.


Adam W. Bargteil is an assistant professor at the University of Utah.  He earned his Ph.D. in computer science from the University of California, Berkeley and spent two years as a post-doctoral fellow in the School of Computer Science at Carnegie Mellon University.  His primary research interests lie in the area of physically based simulation for computer animation.  His research has focused on developing tools that allow animators to create high-quality realistic, visually detailed animations of complex materials.  Adam has co-authored three SIGGRAPH papers and three animated shorts showcasing his projects have appeared in the SIGGRAPH Electronic Theater.  He received dual BS degrees in computer science and mathematics (magna cum laude) from the University of Maryland in 2000.  Adam was a U.C. Microelectronics Fellow in 2000 and a Siebel Scholar in 2006.  From 2005 to 2007, he was a consultant at PDI/DreamWorks, developing fluid simulation tools that were used in "Shrek the Third" and "Bee Movie."

Mary Hall (PhD 1991, Rice University) recently joined the School of Computing at University of Utah as an associate professor.  Her research focuses on compiler technology for exploiting performance-enhancing features of a variety of computer architectures. Prior to joining University of Utah, Prof. Hall held positions at University of Southern California, Caltech, Stanford and Rice University.  Prof. Hall is currently leading the autotuning group in the DOE SciDAC Performance Engineering Research Institute.  Previously, she was principal investigator on DIVA (Data-IntensiVe Architecture), a system architecture project that utilizes processing logic internal to memory chips as smart memory coprocessors, and on DEFACTO (Design Environment for Adaptive Computing), an end-to-end design environment for FPGA-based computing environments.