A Comparison of Penguin Swarm Optimization Algorithms for Enhancing Network Throughput

The increasing integration of information technology (IT) into computer communication systems has resulted in smart grid systems that promise increased efficiency, reliability, and sustainability. Managing and analyzing the vast amounts of data generated by smart grid devices is critical to effectively manipulating the potential of these systems. Conventional data centers have been used to process and store data, however, the emergence of edge computing technologies has made alternative data management, a decentralized approach to shifting data from highly over-loaded virtual machines to lighter-loaded nodes, where the queue task is increased from a certain threshold. With its ability to deliver computing and storage resources at the network's edge, close to device terminals and users, edge computing is an evolving distributed computing approach well-suited for facilitating extensive management of smart devices in the future smart grid. Inspired by the penguin behavior, the Penguin Colony Optimization algorithm (PeCO) is a new meta-heuristic technique used in this study to solve the network load balancing issues. Penguins are alive in extremely frigid climates around the world. A penguin's body heat intensity is the indicator of its fitness and draws other penguins in its vicinity towards it so that they may stay warm collectively. The magnitude of body heat radiation the penguin produces is correlated with its heat strength. The Penguin Search algorithm (PeSO), another algorithm inspired by penguins, has demonstrated vital efficacy in dropping response times and queue lengths on nodes with the highest fitness values. The traffic throughput throughout peak hours was significantly enhanced by 62% when PeSO used the Circle multimodal function, as opposed to 62.37% when PeSO used the Raster-gin multi-modal function. Circle multi-modal function via PeSO of 64.35% yields an enhanced result.