Distributed Deep Learning Inference Acceleration using Seamless Collaboration in Edge Computing

TL;DR

This paper introduces HALP, a novel distributed CNN inference scheme for edge computing that accelerates inference time and improves reliability by optimizing task collaboration and considering receptive fields.

Contribution

The paper proposes HALP, a new task collaboration scheme that enhances distributed CNN inference efficiency and reliability in edge computing environments.

Findings

HALP accelerates VGG-16 inference by 1.7-2.0x on GTX 1080TI and JETSON AGX Xavier.

HALP outperforms state-of-the-art MoDNN in inference speed.

HALP ensures high service reliability under time-variant channel conditions.

Abstract

This paper studies inference acceleration using distributed convolutional neural networks (CNNs) in collaborative edge computing. To ensure inference accuracy in inference task partitioning, we consider the receptive-field when performing segment-based partitioning. To maximize the parallelization between the communication and computing processes, thereby minimizing the total inference time of an inference task, we design a novel task collaboration scheme in which the overlapping zone of the sub-tasks on secondary edge servers (ESs) is executed on the host ES, named as HALP. We further extend HALP to the scenario of multiple tasks. Experimental results show that HALP can accelerate CNN inference in VGG-16 by 1.7-2.0x for a single task and 1.7-1.8x for 4 tasks per batch on GTX 1080TI and JETSON AGX Xavier, which outperforms the state-of-the-art work MoDNN. Moreover, we evaluate the service reliability under time-variant channel, which shows that HALP is an effective solution to ensure high service reliability with strict service deadline.