ScissionLite: Accelerating Distributed Deep Learning With Lightweight Data Compression for IIoT

Industrial Internet of Things (IIoT) applications can benefit from leveraging edge computing.For example, applications relying on deep neural network (DNN) models can be sliced and distributed across the IIoT device and the edge of the network for decreasing the latency of inference.However, low network performance between IIoT devices and the edge is often a bottleneck.In this study, we propose ScissionLite, a holistic framework for accelerating distributed DNN inference using lightweight data compression.For the compression method, we implement a new lightweight down/upsampling network for performance-limited IIoT devices, which is inserted at the slicing point of a DNN model in order to decrease the outbound network traffic without a significant accuracy drop.We also develop a benchmarking tool to accurately find the optimal slicing point of the DNN for the best inference latency.ScissionLite improves the inference latency by up to 15.7x with a minimum accuracy degradation.