High-throughput neural network inference requires coordinating many optimization decisions, including parallel tiling, microkernel selection, and data layout. The product of these decisions forms a search space of programs which is typically intractably large. Existing approaches (e.g., auto-schedulers) often address this problem by sampling this space heuristically. In contrast, we introduce a dynamic-programming-based approach to explore more of the search space by iteratively decomposing large program specifications into smaller specifications reachable from a set of rewrites, then composing a final program from each rewrite that minimizes an affine cost model. To reduce memory requirements, we employ a novel memoization table representation, which indexes specifications by coordinates in and compresses identical, adjacent solutions. This approach can visit a much larger set of programs than prior work. To evaluate the approach, we developed Morello, a compiler which lowers specifications roughly equivalent to a few-node XLA computation graph to x86. Notably, we found that an affine cost model is sufficient to surface high-throughput programs. For example, Morello synthesized a collection of matrix multiplication benchmarks targeting a Zen 1 CPU, including a 1x2048x16384, bfloat16-to-float32 vector-matrix multiply, which was integrated into Google'sthis http URL.
View on arXiv@article{kaufman2025_2505.01637,
title={ Morello: Compiling Fast Neural Networks with Dynamic Programming and Spatial Compression },
author={ Samuel J. Kaufman and René Just and Rastislav Bodik },
journal={arXiv preprint arXiv:2505.01637},
year={ 2025 }
}