A study of workflow task scheduling strategy for makespan and cost optimization in jointcloud computing

As an emerging cloud computing model, jointcloud computing significantly improves resource utilization and system elasticity by supporting open collaboration among cloud service providers and deep integration of multi-party resources, providing users with a more personalized service experience. However, in the inter-cloud computing environment, the workflow task scheduling problem faces a number of challenges, including the mapping of cross-cloud resources to tasks, the determination of the execution order of subtasks, the satisfaction of priority constraints, and the guarantee of service quality requirements. In this paper, we address this problem by proposing a task scheduling algorithm with a hybrid genetic algorithm and ant colony algorithm, aiming to optimize the completion time and execution cost of workflow tasks in inter-cloud computing environments. Experimental results show that the algorithm significantly reduces the makesapn and achieves a better balance in execution cost compared to the other algorithms in workflow task scenarios of different sizes.