基于随机服务决策网的云计算速度调节动态优化技术研究

Dynamic optimization of speed scaling in cloud computing based on stochastic service decision nets

通过在不同的状态下进行速度调节能够达到降低能耗的目的,研究了基于随机服务决策网模型的云计算速度动态优化的技术。这种模型将随机Petri网络和Markov决策过程模型结合在一起,从而能够动态调整速度调节策略和进行性能评估。同时这种模型是以服务为导向的,因此能够运用典型模式将复杂模型简化为具有较小状态空间的简单模型。这种模型能够描述复杂的系统行为和决策过程,仿真也表明了这种模型的有效性。

By shifting among these states using speed scaling, the energy consumption was proportional to the workload, which was termed energy proportionality. This study used stochastic service decision nets to investigate energy-efficient speed scaling on Web servers. This model combined stochastic Petri nets with Markov decision process models. It enabled the model to dynamically optimize the speed scaling strategy and make performance evaluations. The model was graphical and intuitive enough to characterize complicated system behavior and decisions. The model was service oriented using the typical service patterns to reduce the complex model to a simple model with a smaller state space. Performance and reward equivalent analyse substantially reduced the system behavior sub-net. The model gave the optimal strategy and evaluated performance and energy metrics more concisely.