Bifrost: End-to-End Evaluation and Optimization of Reconfigurable DNN Accelerators
This work addresses a barrier for researchers in accelerator design by automating evaluation and optimization, though it is incremental as it builds on existing tools like STONNE and TVM.
The paper tackles the manual and time-consuming process of evaluating and optimizing reconfigurable DNN inference accelerators by introducing Bifrost, an end-to-end framework that automates model preparation and configuration exploration, achieving speedups of 50x for convolutional layers and 11x for fully connected layers in AlexNet.
Reconfigurable accelerators for deep neural networks (DNNs) promise to improve performance such as inference latency. STONNE is the first cycle-accurate simulator for reconfigurable DNN inference accelerators which allows for the exploration of accelerator designs and configuration space. However, preparing models for evaluation and exploring configuration space in STONNE is a manual developer-timeconsuming process, which is a barrier for research. This paper introduces Bifrost, an end-to-end framework for the evaluation and optimization of reconfigurable DNN inference accelerators. Bifrost operates as a frontend for STONNE and leverages the TVM deep learning compiler stack to parse models and automate offloading of accelerated computations. We discuss Bifrost's advantages over STONNE and other tools, and evaluate the MAERI and SIGMA architectures using Bifrost. Additionally, Bifrost introduces a module leveraging AutoTVM to efficiently explore accelerator designs and dataflow mapping space to optimize performance. This is demonstrated by tuning the MAERI architecture and generating efficient dataflow mappings for AlexNet, obtaining an average speedup of $50\times$ for the convolutional layers and $11\times$ for the fully connected layers. Our code is available at www.github.com/gicLAB/bifrost.