An Effective Framework of Program Optimization for High Performance Computing

The increasing complexity of modern architectures and memory models challenges the design of optimizing compilers. It is mandatory to perform several optimizing transformations of the original program to exploit the machine to its best, especially for scientific, computational intensive codes. Aiming at investigating the best transformation sequence and the best transformation parameters simultaneously, this paper combines polyhedral model and empirical search to create a powerful optimization framework that is capable of fully automated non-parametric transformations and automatic parameter search. The framework employs polyhedral model to facilitate the search of non-parametric code transformation composition, and designs uses Genetic Algorithms to find the optimal parameters. The framework is demonstrated on three typical computational kernels for code transformations to achieve performance that greatly exceeds the native compiler, and is significantly better than state-of-the-art polyhedral model based loop transformations and iterative compilation, generating efficient code on complex loop nests.