A Comparative Study of Energy Consumption in Nature-Inspired Optimization Algorithms

The problem of optimisation is a fundamental mathematical challenge encountered in various fields to determine the optimal solution to a problem while adhering to a set of constraints. Nature-inspired optimisation algorithms are a novel category of algorithms that take inspiration from natural systems to solve complex optimization problems. The utilisation of Nature-Inspired Optimisation (NIO) algorithms has gained widespread popularity in addressing a diverse range of optimisation problems in practical settings due to their straightforwardness, adaptability, and efficacy. NIO algorithms have been applied in diverse areas of ICT, such as metaverse and cloud computing, with a focus on virtual machines, storage, security, meta-analytics, and other related domains. This study presents a comparative analysis of energy consumption and associated carbon footprint of four widely used NIO algorithms. Intel Power Gadget was used to quantify the energy utilisation throughout the execution of each algorithm. Additionally, the carbon footprint of each algorithm is determined by referencing the UK DEFRA guide. The findings of this investigation indicate that different algorithms exhibit varying patterns of energy consumption to attain a common objective. Furthermore, a t-test was performed, indicating that the mean energy consumption of each algorithm exhibits significant differences from one another. Regarding prospective research it is worth considering various CPU architectures, including the Apple bionic chipset and the latest Intel processors, as potential options for executing these algorithms. The collection of energy consumption data for various CPU architectures can facilitate the identification of potential correlations between NIO algorithms and hardware resource energy consumption.