Modified tabu-based ant colony optimisation algorithm for energy-efficient cloud computing systems

The widespread adoption of Cloud Computing (CC) and rapid rise in capacity and scale of data centres to host and provide services via internet results in a significant increase in electricity usage and increased carbon footprints. One of the challenging research problems is to motivate green CC by reducing datacentres energy consumption. The well-known Task Scheduling (TS) considered as NP-hard problem needs to be addressed to facilitate green CC which influences the overall efficiency of cloud system. This paper presented a modified novel Tabu-based Ant Colony Optimisation Algorithm (TACO) energy efficient TS approach for heterogeneous cloud environment. TACO maintains a Long-Term Memory (LTM) in terms of aspiration list and Short-Term Memory (STM) in terms of tabu list. TACO is evaluated in CloudSim 3.0.3 to measure its performance with existing energy-based metaheuristics Particle Swarm Optimisation (PSO) and Genetic Algorithm (GA) which shows its outperformance in energy consumption, makespan and resource utilisation.