Mar 22, 2010

Yi and Whaley receive $360K DOE grant




The Department of Energy (DOE) Office of Science has awarded a $360,000 grant to support CS assistant professors Qing Yi and Clint Whaley's research developing an environment for programmable code optimization and empirical tuning. This research will be conducted in collaboration with Daniel Quinlan at the Lawrence Livermore National Laboratory.

In this research, the investigators will build an integrated environment for improving application performance through programmable code optimization and empirical tuning. This optimization environment will support applications in multiple programming languages, including C, C++, and Fortran 2003. The approach explored in this research will permit different levels of automation, from semi-automated performance optimization to fully-automated tuning. This flexibility in the environment aims to maximize the performance improvement for a given amount of developer interaction. It will, in this way, provide a framework that scales from thorough optimization of heavily-used computational kernels in domain specific libraries to tuning of whole programs.

For more information, see the project webpage.