MFP: an approach to delay and energy-efficient module placement in IoT applications based on multi-fog
- PDF / 1,342,316 Bytes
- 17 Pages / 595.276 x 790.866 pts Page_size
- 54 Downloads / 153 Views
ORIGINAL RESEARCH
MFP: an approach to delay and energy-efficient module placement in IoT applications based on multi-fog Morteza Dadashi Gavaber1 · Amir Rajabzadeh1 Received: 10 August 2019 / Accepted: 4 September 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract One of the challenges of using fog computing in IoT systems is the efficient placement of resources in IoT applications. This paper presents a resource placement method for fog-based IoT systems to reduce their latency and energy consumption. Given the limited processing power of fog nodes, only a limited number of modules can be run on these nodes. In fog-cloud systems, placing the modules on fog nodes instead of the cloud layer can be expected to reduce system latency. Therefore, to achieve enhanced latency and energy consumption, this paper introduces a multi-zone fog layer architecture where each zone is a multi-fog. The core idea of the proposal is to use the idle processing capacity of fog nodes in each zone through the maximal placement of modules on these nodes. The paper also presents an algorithm called MFP for carrying out this placement. To evaluate the proposed algorithm, it was simulated in iFogSim for two scenarios with different topologies. The simulation results showed that the proposed scheme offers 16.81% lower latency and 17.75% lower energy consumption than the existing schemes. Keywords Fog computing · Internet of Things · Multi-fog · Resource placement · Energy-efficient · Delay-efficient
1 Introduction With the dramatic advancement of communications technologies in recent years, the Internet of Things (IoT) has emerged as a paradigm for the next generation of computing (Gubbi et al. 2013). It has been estimated that by 2020, over 50 billion IoT devices will be in use for a wide variety of applications across the world (Abdul-Qawy and Srinivasulu 2019). The standard infrastructure for the implementation of IoT applications is cloud (Boveiri et al. 2018). As illustrated in Fig. 1a, cloud-based IoT structures consist of two tiers (Li et al. 2017): (1) the physical IoT components that perform the sensing/actuating functions, and (2) the cloud that is responsible for controlling and providing the services required by IoT devices (Li et al. 2017). Although the combination of cloud and IoT offers many benefits including low-cost installation and integration for * Amir Rajabzadeh [email protected] Morteza Dadashi Gavaber [email protected] 1
Department of Computer Engineering and Information Technology, Razi University, Kermanshah, Iran
complex data processing and deployment, such systems are ill-suited for real-time, delay-sensitive, and energy-efficient services (Mahmoud et al. 2018). To address these challenges, Cisco has introduced a new technology called fog computing (Mahmoud et al. 2018; Wang et al. 2017). Figure 1b shows the architecture of a fog computing-based IoT system comprised of three tiers: a layer of physical components or IoT tier, a second layer for fog computing, and the third l
Data Loading...
