![]()
Mary Lou Soffa Developing a Foundation for Code Optimization
Although code optimization technology has been successful over the past 40 years, recent trends are emerging that demand we reconsider the paradigm that we are using for code optimization. In particular, the trends toward dynamic optimization, writing embedded system software in high level languages and the lack of significant performance improvement from current optimization research are forcing us to rethink what we know and do not know about optimization. The challenge in optimization today is to explore properties of optimizations and develop a framework for better understanding and use of optimizations. This talk will discuss the problems and then present a framework for predicting the profitability of optimizations. The prediction is done using models that we developed for code optimizations, the code context where an optimization is to be applied and the resources available. We experimentally used the models to selectively determine what optimization to apply and for determining the application order of optimizations to improve performance. Results will be presented that demonstrates our approach is effective and has a much lower overhead than current approaches for determining the order of optimizations. Short Bio Mary Lou Soffa is the Owen R. Cheatham Professor and Chair of the Computer Science
Department at the University of Virginia. She previously was Professor of
Computer Science at the University of Pittsburgh. Her research interests include
software engineering, program analysis, optimizing compilers, and software tools
for debugging and testing programs. In 1999, she received the Presidential Award
for Excellence in Science, Mathematics and Engineering Mentoring, given by the
Back to Distinguished Lecture Series - Fall 2004 / Spring 2005 Schedule |
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||