Efficient search‐space pruning for integrated fusion and tiling transformations

Abstract

Compile‐time optimizations involve a number of transformations such as loop permutation, fusion, tiling, array contraction etc. The selection of the appropriate transformation to minimize the execution time is a challenging task. We address this problem in the context of tensor contraction expressions involving arrays too large to fit in main memory. Domain‐specific features of the computation are exploited to develop an integrated framework that facilitates the exploration of the entire search space of optimizations. In this paper, we discuss the exploration of the space of loop fusion and tiling transformations in order to minimize the disk I/O cost. These two transformations are integrated and pruning strategies are presented that significantly reduce the number of loop structures to be evaluated for subsequent transformations. The evaluation of the framework using representative contraction expressions from quantum chemistry shows a dramatic reduction in the size of the search space using the strategies presented. Copyright © 2007 John Wiley & Sons, Ltd.

Special Issue: Current Trends in Compilers for Parallel Computers

“…As for Gao et al [7], they exploit in their paper domain-specific features of the tensor contraction expressions to guide memory hierarchy optimizations. Specifically, such knowledge is used to develop an integrated framework that facilitates the exploration of the search space for the fusion and tiling transformations in order to minimize disk accesses when such expressions involve arrays too large to fit in main memory.…”