Abstract: It proposes a static forecast method on the basis of the change of parameter to forecast execution time of programs, according to the number of code executed. It obtains the number of code executed by analysing source code of task and resource evaluation, draws the trend of number of code executed changed with parameter by automatic curve fitting, finds out the relationship between the number of code executed and running time by operating applications practically, and program running time with the tendency of changes in parameters is obtained. It uses PASCAL Hanoi program to verify this forecast method, measures the program run time with the change of parameters by fitting test and running program. The results show that the predicted run time is very close to the real running time.

Key words: parameters change, execution time prediction, cloud application, static forecast

摘要： 根据程序执行的语句条数，提出了一种基于参数变化的程序执行时间预估方法。通过分析任务源码，进行资源评估，得出任务执行的语句数；通过自动曲线拟合，得出语句数随参数变化的趋势；通过实际运行程序，找出了语句数与运行时间的关系，得出程序运行时间随参数变化的趋势。使用pascal汉诺塔程序对预估方法进行验证，发现测得的参数变化下的程序运行时间与实际运行时间非常接近，误差率很小。

关键词: 参数变化, 执行时间预测, 云应用程序, 静态预估