针对用户、服务提供商(SP)和基础设施提供商三者之间的市场行为所带来的收益问题,该文面向服务提供商提出了一种两阶段多虚拟机资源分配机制。首先,建立能力规划模型并引入粒子群算法,决策出服务提供商收益最大化的资源购买方案。然后,提出一种优化用户满意度的服务提供策略,优化了服务提供商的长期收益。最后仿真实验结果表明,该方法有效提高了服务提供商的收益,并取得了较好的用户满意度。
Considering Service Providers (SP) revenue optimization issue, a two-stage multi-virtual machine resource provisioning mechanism is proposed in this paper. Firstly, a capacity planning model is proposed and the particle swarm algorithm is used to make resource purchase set, which can gain the maximize profit for SP. Then, a utility function to measure customer's satisfaction is proposed to optimize SP's profit in the long way. Simulation experimental results show that the proposed method improves effectively SP's profit, and gains better user's satisfaction.