With the rapid development of information and communication technology, cloud storage systems are facing huge energy consumption problems while carrying massive data processing. In order to alleviate these pressures and meet the diverse needs of user groups, this paper establishes a partial server synchronous multiple vacation \({\text{M/M/}}c + k\) queuing system with preemptive priority and repairable work breakdowns. The virtual machine (VM) scheduling strategy can be applied to improve resource utilization. By analyzing the energy-saving effects of the model, the results show that the system’s energy conservation increases by a factor 2.08 when the service rate in Zone II increases from 2.2 to 3.0. So regardless of whether member tasks arrive at the service system at a low rate or a high rate, optimizing the service rate and reasonable VM resource scheduling can significantly reduce the energy consumption of the cloud storage system. In addition, this paper constructs individual income and social income functions respectively, utilizing the Seagull Optimization Algorithm (SOA) to obtain the socially optimal arrival rate of member tasks. Based on this, a reasonable pricing strategy is formulated. Revenue analysis suggests that cloud service providers (CSPs) should monitor request tasks within the system in real time to avoid reducing the enthusiasm of users to join the team due to excessive congestion, so as to achieve a balance between system performance optimization and energy consumption control.
