A hybrid DECSO algorithm for efficient multi objective task scheduling in cloud computing environments

Abstract One of the biggest problems in the rapidly developing cloud computing field in recent years is efficient task scheduling. Task scheduling in cloud computing is recognized as an NP-complete problem, presenting significant challenges due to the large task sizes and the complexity of efficiently managing diverse computational resources. Task scheduling in cloud computing aims to ensure that tasks are assigned to virtual machines to minimize completion time and maximize resource utilization. To address these challenges, this study introduces a novel hybrid optimization algorithm named Differential Evolution Cat Swarm Optimization (DECSO). Unlike traditional hybrid approaches, DECSO dynamically balances exploration and exploitation, ensuring a more adaptive and efficient task scheduling strategy. DECSO synergizes the global exploration ability and adaptive capabilities of Differential Evolution (DE) with the local search efficiency and explorative and exploitative strengths of Cat Swarm Optimization (CSO). The proposed DECSO algorithm is compared with PSO (Particle Swarm Optimization) and CSO via the CloudSim simulation experiment platform. DECSO’s performance is evaluated using makespan, resource utilization, and migration time, which are critical metrics for efficient cloud task scheduling. The experimental results demonstrate that DECSO achieves up to 22.6% reduction in MakeSpan compared to CSO and 9.6% compared to PSO, 11.9% improvement in resource utilization compared to CSO and 14.7% compared to PSO, and 20.6% reduction in migration time compared to CSO and 11.2% compared to PSO. The results obtained from the simulation studies carried out show that the presented optimization model provides significant improvements in terms of MakeSpan, resource utilization, and migration time.