Array Unification: A Locality Optimization Technique

Abstract

One of the key challenges facing computer architects and compiler writers is the increasing discrepancy between processor cycle times and main memory access times. To alleviate this problem for a class of array-dominated codes, compilers may employ either control-centric transformations that change data access patterns of nested loops or data-centric transformations that modify the memory layouts of multi-dimensional arrays. Most of the layout optimizations proposed so far either modify the layout of each array independently or are based on explicit data reorganizations at runtime. This paper describes a compiler technique, called array unification, that automatically maps multiple arrays into a single data (array) space to improve data locality. We present a mathematical framework that enables us to systematically derive suitable mappings for a given program. The framework divides the arrays accessed by the program into several groups and each group is transformed to improve spatial locality and reduce the number of conflict misses. As compared to the previous approaches, the proposed technique works on a larger scope and makes use of independent layout transformations as well whenever necessary. Preliminary results on two benchmark codes show significant improvements in cache miss rates and execution time.