Guardians of the Deep Fog: Failure-Resilient DNN Inference from Edge to Cloud

TL;DR

This paper presents deepFogGuard, a novel DNN architecture augmentation that enhances failure resilience in distributed edge-cloud inference by introducing skip hyperconnections inspired by residual networks.

Contribution

We propose deepFogGuard, a new scheme that augments distributed DNNs with skip hyperconnections to improve failure resilience in edge-cloud environments.

Findings

deepFogGuard improves inference resilience in distributed DNNs

Experimental results confirm effectiveness on sensing and vision datasets

Resiliency is achieved without significant performance loss

Abstract

Partitioning and distributing deep neural networks (DNNs) over physical nodes such as edge, fog, or cloud nodes, could enhance sensor fusion, and reduce bandwidth and inference latency. However, when a DNN is distributed over physical nodes, failure of the physical nodes causes the failure of the DNN units that are placed on these nodes. The performance of the inference task will be unpredictable, and most likely, poor, if the distributed DNN is not specifically designed and properly trained for failures. Motivated by this, we introduce deepFogGuard, a DNN architecture augmentation scheme for making the distributed DNN inference task failure-resilient. To articulate deepFogGuard, we introduce the elements and a model for the resiliency of distributed DNN inference. Inspired by the concept of residual connections in DNNs, we introduce skip hyperconnections in distributed DNNs, which are the basis of deepFogGuard's design to provide resiliency. Next, our extensive experiments using two existing datasets for the sensing and vision applications confirm the ability of deepFogGuard to provide resiliency for distributed DNNs in edge-cloud networks.