Equinox: Training (for Free) on a Custom Inference Accelerator
Mario Paulo Drumond, Louis Coulon, Arash Pourhabibi, Ahmet Caner Yüzügüler, Babak Falsafi, Martin Jaggi · 2021
DNN inference accelerators executing online services exhibit low average loads because of service demand variability, leading to poor resource utilization. Unfortunately, reclaiming idle inference cycles is difficult as other workloads can not execute on a custom accelerator. With recent proposals for the use of fixed-point arithmetic in training, there are opportunities for training services to piggyback on inference accelerators. We make the observation that a key challenge in doing so is maintaining service-level latency constraints for inference. We show that relaxing latency constraints in an inference accelerator with ALU arrays that are batching-optimized achieves near-optimal throughput for a given area and power envelope while maintaining inference services’ tail latency goals.