Optimus: towards optimal layer-fusion on deep learning processors

Xuyi Cai, Ying Wang, Lei Zhang

2021 · 13 citations · 38 references

DOIFull text

Open access

Concepts

Abstract

Neural network layer fusion has been proposed to parallelize the inference of neural layers and thus significantly reduces the feature-induced memory accesses. However, how to fuse the neural layers is still a challenging issue that heavily depends on both the network architecture and the specific DNN processor configuration. This work formalizes the layer fusion problem for DNN processors, proves that prior fusion solutions cannot guarantee memory-level optimality, and presents a novel neural network fusion framework, Optimus. Optimus includes an accurate memory cost model to evaluate fusion schemes, and a Computing-Graph (CG) based layer fusion algorithm, which generates high-efficiency layer-fusion schemes for arbitrary network architectures on DNN processors. The proposed off-line and on-line graph-based fusion algorithms can reduce 10.1% - 72.2% off-chip memory traffic and obtain 1.71x - 3.94x energy efficiency over SOTA baselines on DNN workloads, and they bring significant power-efficiency boost to the DNN processors of different architectures and dataflows.

References

38