云计算环境下基于热力学演化算法的任务调度与虚拟机分配算法研究

Task Scheduling and Virtual Machine Allocation Strategy Based on Thermodynamics Evolutionary Algorithm in Cloud Computing Environment

在云计算环境下的云任务调度和虚拟机分配过程建立了数学模型,并将其转换为整数编码形式的组合优化问题,并提出了一种热力学演化算法进行问题求解。算法根据整数编码形式定义了基因熵和个体能量,并引入了温度的概念,算法中提出了两种选择策略,算法利用自由能极小值原理驱动种群向最优化方向演化。实验结果表明热力学演化算法可以有效地解决云任务调度和虚拟机分配问题,可以为云环境调度问题提供依据。

The mathematical model of cloud task scheduling and virtual machine allocation process in cloud computing cloud environment was established. And the model was transformed to the combination optimization problem with the integer encoding. The thermodynamic evolution algorithm was proposed for problem solving. In the algorithm, the gene entropy and individual energy were defined according to the integer encoding, and the concept of the temperature was introduced. Two selection strategies were designed in the thermodynamic evolution algorithm and the free energy minimum principle was used to drive the population evolution for the optimization direction. Experimental results show that the thermodynamics evolutionary algorithm can effectively solve the problem of the cloud task scheduling and the virtual machine allocation and can provide basis for the cloud environment scheduling problem.