FlexPie: Accelerate Distributed Inference on Edge Devices with Flexible Combinatorial Optimization[Technical Report]

TL;DR

FlexPie introduces a novel approach to accelerate distributed inference on edge devices by leveraging flexible combinatorial optimization, addressing the challenges of deploying deep neural networks in IoT environments.

Contribution

The paper proposes a new optimization framework that enhances the efficiency of collaborative inference across multiple edge devices.

Findings

Significant reduction in inference latency on edge networks

Improved resource utilization across distributed devices

Enhanced scalability for IoT applications

Abstract

The rapid advancement of deep learning has catalyzed the development of novel IoT applications, which often deploy pre-trained deep neural network (DNN) models across multiple edge devices for collaborative inference.