Data-Centric Transformations for Locality Enhancement

摘要

On modern computers, the performance of programs is often limited by memory latency rather than by processor cycle time. To reduce the impact of memory latency, the restructuring compiler community has developed locality-enhancing program transformations such as loop permutation and tiling. These transfor- mations work well for perfectly nested loops (loops in which all assignment statements are contained in the innermost loop), but their performance on codes such as matrix factorizations that contain imperfectly nested loops leaves much to be desired.