服务延迟保障的混合绿色云数据中心成本最小化方法

With the rapid development of cloud computing, more and more companies deploy and run a growing number of heterogeneous applications to a private CDC. The number and types of applications have greatly increased the operating cost of private CDC providers, and the number of users has increased greatly. To effectively reduce the operating costs of CDC providers while meeting application performance, aiming at the intelligent task scheduling method of hybrid green cloud environment, a Cost Minimization Algorithm (CMA) in hybrid green CDC was proposed. According to the temporal differences between energy consumption of private CDC, available green energy, execution prices of public clouds and so on, the task scheduling provided by CMA could intelligently schedule all arriving tasks to be executed in private CDC and public clouds, and the service delay bounds of tasks were strictly guaranteed. On this basis, a genetic learning particle swarm optimization was proposed. Real-life data-driven experimental results demonstrated that the proposed method significantly reduced the cost of private CDC providers compared with existing typical algorithms.