Communication-Optimal Tilings for Projective Nested Loops with Arbitrary Bounds
share
Reducing communication - either between levels of a memory hierarchy or between processors over a network - is a key component of performance optimization (in both time and energy) for many problems, including dense linear algebra, particle interactions, and machine learning. For these problems, which can be represented as nested-loop computations, previous tiling based approaches have been used to find both lower bounds on the communication required to execute them and optimal rearrangements, or blockings, to attain such lower bounds. However, such general approaches have typically assumed the problem sizes are large, an assumption that is often not met in practice. For instance, the classical (# arithmetic operations)/(cache size)^1/2 lower bound for matrix multiplication is not tight for matrix-vector multiplications, which must read in at least O(# arithmetic operations) words of memory; similar issues occur for almost all convolutions in machine learning applications, which use extremely small filter sizes (and therefore, loop bounds). In this paper, we provide an efficient way to both find and obtain, via an appropriate, efficiently constructible blocking, communication lower bounds and matching tilings which attain these lower bounds for nested loop programs with arbitrary loop bounds that operate on multidimensional arrays in the projective case, where the array indices are subsets of the loop indices. Our approach works on all such problems, regardless of dimensionality, size, memory access patterns, or number of arrays, and directly applies to (among other examples) matrix multiplication and similar dense linear algebra operations, tensor contractions, n-body pairwise interactions, pointwise convolutions, and fully connected layers.
READ FULL TEXT