English

Sawtooth Wavefront Reordering: Enhanced CuTile FlashAttention on NVIDIA GB10

Performance 2026-01-27 v2 Artificial Intelligence Machine Learning Operating Systems

Abstract

High-performance attention kernels are essential for Large Language Models. This paper presents analysis of CuTile-based Flash Attention memory behavior and a technique to improve its cache performance. In particular, our analysis on the NVIDIA GB10 (Grace Blackwell) identifies the main cause of L2 cache miss. Leveraging this insight, we introduce a new programming technique called Sawtooth Wavefront Reordering that reduces L2 misses. We validate it in both CUDA and CuTile, observing 50\% or greater reduction in L2 misses and up to 60\% increase in throughput on GB10.

Cite

@article{arxiv.2601.16032,
  title  = {Sawtooth Wavefront Reordering: Enhanced CuTile FlashAttention on NVIDIA GB10},
  author = {Yifan Zhu and Yekai Pan and Chen Ding},
  journal= {arXiv preprint arXiv:2601.16032},
  year   = {2026}
}
R2 v1 2026-07-01T09:15:56.658Z