Optimising Cloud Servers Utilisation Based on Locust-Inspired Algorithm

The efficient management of resource sharing plays a crucial role in the cloud execution environment. Cloud computing presents this contribution by providing solid infrastructure services on a pay-as-you-go basis with assuring service availability and high service quality to meet the user needs. Cloud services are provided through various data centres that are separated around the world whereas the users have been served throughout these data centres which are represented as resources; to efficiently utilise these resources, the virtualisation is introduced as Virtual Machines (VMs). Therefore, resource allocation plays a vital role in utilising data centres perfectly. At the same time, the VMs migration from physical machine to another could attain maximum utilisation of the physical machine (server). Hence, the previous inspired algorithms could not exploit the servers efficiently. This paper introduces Locust-Inspired optimisation Algorithm (LIOA) which is an evolutionary optimisation algorithm that is used to find an optimised solution to utilise the servers efficiently. Comparing to the previous VMs migration and allocation algorithms such as Static Threshold with Minimum Migration Time (THR-MMT) and Locust-inspired scheduling Algorithm to reduce Cloud computing consumed Energy (LACE). Our algorithm presented a high probability to find the global best solution. To do so, the proposed algorithm was evaluated with the CloudSim toolkit within a real data-set. This work was compared with an existing heuristic algorithm (THR-MMT) and metaheuristic (LACE) in which LIOA outperforms the algorithms by significant enhancements in the server utilisation.