Energy Efficient Service Distribution in Internet of Things

TL;DR

This paper proposes an optimized multi-layer IoT architecture with fog and cloud computing to significantly reduce power consumption by enabling local computation at the IoT layer.

Contribution

It introduces a method for optimizing service placement in a multi-layer IoT architecture to minimize power use, highlighting the benefits of local computation.

Findings

Local computation can save up to 90% power compared to central cloud servers.

Optimized service placement reduces overall power consumption in IoT networks.

Multi-layer architecture enhances energy efficiency in IoT deployments.

Abstract

The Internet of Things (IoT) networks are expected to involve myriad of devices, ranging from simple sensors to powerful single board computers and smart phones. The great advancement in computational power of embedded technologies have enabled the integration of these devices into the IoT network, allowing for cloud functionalities to be extended near to the source of data. In this paper we study a multi-layer distributed IoT architecture supported by fog and cloud. We optimize the placement of the IoT services in this architecture so that the total power consumption is minimized. Our results show that, introducing local computation at the IoT layer can bring up to 90% power savings compared with general purpose servers in a central cloud.