Abstract

Using static analysis techniques compilers for lazy functional languages can identify parts of a program that can be legitimately evaluated in parallel with the main thread of execution. These techniques can produce improvements in the runtime performance of a program, but are limited by the static analyses’ poor prediction of runtime performance. This talk outlines the development of a system that uses iterative compilation in addition to well-studied static analysis techniques. Our representation of the parallel programs allows us to use traditional ‘blind’ search techniques or profile-directed improvement. We compare the results of different search strategies and discuss the pitfalls and limitations of our technique. Overall, the use of iterative feedback allows us to achieve higher performance than through static analysis alone.