A Residual Resource Fitness-Based Genetic Algorithm for a Fog-Level Virtual Machine Placement for Green Smart City Services

Sanjoy Choudhury, Ashish Kumar Luhach (), Joel J. P. C. Rodrigues, Mohammed AL-Numay, Uttam Ghosh and Diptendu Sinha Roy
Additional contact information
Sanjoy Choudhury: Department of Computer Science and Engineering, National Institute of Technology Meghalaya, Shillong 793003, India
Ashish Kumar Luhach: Department of Electrical and Communication Engineering, The PNG University of Technology, Lae MP 411, Papua New Guinea
Joel J. P. C. Rodrigues: College of Computer Science and Technology, China University of Petroleum (East China), Qingdao 266555, China
Mohammed AL-Numay: Electrical Engineering Department, College of Engineering, King Saud University, P.O. Box 800, Riyadh 11421, Saudi Arabia
Uttam Ghosh: Department of Computer Science and Data Science, Meharry Medical College, Nashville, TN 37208, USA
Diptendu Sinha Roy: Department of Computer Science and Engineering, National Institute of Technology Meghalaya, Shillong 793003, India

Sustainability, 2023, vol. 15, issue 11, 1-21

Abstract: Energy efficient information and communication technology (ICT) infrastructure at all levels of a city’s edifice constitutes a core requirement within the sustainable development goals. The ICT infrastructure of smart cities can be considered in three levels, namely the cloud layer infrastructure, devices/sensing layer infrastructure, and fog layer infrastructure at the edge of the network. Efficiency of a data-centre’s energy infrastructure is significantly affected by the placement of virtual machines (VMs) within the data-centre facility. This research establishes the virtual machine (VM) placement problem as an optimisation problem, and due to its adaptability for such complicated search issues, this paper applies the genetic algorithm (GA) towards the VM placement problem solution. When allocating or reallocating a VM, there is a large quantity of unused resources that might be used, however these resources are inefficiently spread over several different active physical machines (PMs). This study aims to increase the data-centre’s efficiency in terms of both energy usage and time spent on maintenance, and introduces a novel fitness function to streamline the process of computing the fitness function in GAs, which is the most computationally intensive component in a GA. A standard GA and first fit decreasing GA (FFD-GA) are applied on benchmark datasets to compare their relative performances. Experimental results obtained using data from Google data-centres demonstrate that the proposed FFD-GA saves around 8% more energy than a standard GA while reducing the computational overhead by approximately 66%.

Keywords: fog computing; virtual machines; genetic algorithm; fitness function; energy efficiency (search for similar items in EconPapers)
JEL-codes: O13 Q Q0 Q2 Q3 Q5 Q56 (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2071-1050/15/11/8918/pdf (application/pdf)
https://www.mdpi.com/2071-1050/15/11/8918/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jsusta:v:15:y:2023:i:11:p:8918-:d:1161461

Access Statistics for this article

Sustainability is currently edited by Ms. Alexandra Wu

More articles in Sustainability from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().