Optimizing cloud resource allocation using Gaussian convergence squirrel search algorithm

Abstract

Cloud computing has gained significant attention for efficiently executing requests according to users’ needs, providing quality services, and optimizing task execution time across virtual machines (VMs). However, resource allocation for various applications is difficult due to dynamic workload conditions and uncertainty in cloud computing. In this research, a Gaussian-based convergence factor (GCF) with the squirrel search algorithm (SSA) is used for allocating resources in a cloud environment for a small finance organization. The proposed GCF along with SSA balances workload and allocates many relevant resources to users’ applications while also ensuring deadline constraints. Tent chaotic map and GCF are incorporated in a traditional SSA to improve the search ability and convergence rate of SSA to allocate optimal resources in the cloud for a small finance organization. The selection of appropriate instance types that align with financial organization requirements includes choosing between on-demand, reserved, or spot instances. The GCF with SSA obtained less energy of 0.505 J, lower execution time of 0.472 s, less makespan of 0.723 s, and higher resource utilization of 51% for 100 tasks of 30 VMs, when compared to existing methods.