OLOA: Based task scheduling in heterogeneous clouds

Abstract

Cloud computing provides increased performance, scalability and is both cost efficient as well as low accounts for maintenance, which makes it a preferred choice when the dynamic allocation of resources is considered. Of the various advantages that cloud computing provides, task scheduling is an essential feature that helps to boost performance and reduce operation cost. In the proposed OLOA, a solution is provided for optimization, taking the makespan and cost as major constraints. This is accomplished using the two algorithms, Lion optimization algorithm (LOA) and the Opposition Based Learning (OBL) algorithm; and creating a hybrid Oppositional Lion optimization algorithm (OLOA). The given solution is simulated and demonstrated in the cloudsim programming environment, where the obtained results show drastic improvement in performance, in comparison to that of the previously used other existing algorithms such as Particle Swarm Optimization (PSO) algorithm, oppositional learning based grey wolf optimizer (OGWO) and the Genetic algorithm (GA), all of which do not match the performance rates of the proposed hybrid algorithm.