基于多线程的多目标任务优先级排序方法仿真  

Simulation of Multi-Objective Task Priority Sorting Method Based on Multithreading

作　　者：黄小平 谢蓉蓉[2] HUANG Xiao-ping;XIE Rong-rong(Guangzhou Institute of Software Department of Network Technology,Guangzhou Guangdong 510990,China;School of Computer Science,Xi'an Shiyou University,Xi^nShanxi 710065,China)

机构地区：[1]广州软件学院网络技术系,广东广州510990 [2]西安石油大学计算机学院,陕西西安710065

出　　处：《计算机仿真》2021年第8期460-464,共5页Computer Simulation

摘　　要：为了提高多目标任务优先级排序效率,降低软件测试的成本,满足软件测试需求,提出基于多线程的多目标任务优先级排序方法,通过多线程并发技术提高多目标演化算法的执行效率,采用融合选择函数的非支配遗传算法,设计编码、算子、个体评价,将平均故障检测率、有效执行时间设为优化目标,实现多目标任务优先级排序。实验结果表明,上述方法的多目标任务优先级排序效果最佳,收敛性好,求解目标函数的速度快,目标函数求解效率高达98%,计算行时间低至1.215ms;平均故障检测率为95%,多目标任务优先级排序准确率高达96%,平均软件测试成本为11.5万元,可降低软件测试成本,提升软件测试效率。In order to improve the efficiency of multi-objective task prioritization,reduce the cost of software testing and meet the requirements of software testing,we reported a multi-objective task prioritization method based on multithreading.Multithreading concurrency technology was applied to improve the efficiency of multi-objective evolutionary algorithm.Through the non-dominated genetic algorithm with selection function,the coding,operator and individual evaluation were designed.The average fault detection rate and effective execution time were set as the optimization objectives.The priority of multi-objective tasks was achieved.The experimental results show that the method has the best priority ranking effect,good convergence,high speed of solving the objective function,the efficiency of solving the objective function is as high as 98%,and the line time is as low as 1.215 ms;The average fault detection rate is 95%,the accuracy of multi-objective task priority is 96%,and the average cost of software testing is 115000 yuan,which can reduce the cost of software testing and improve the efficiency of software testing.

关 键 词：多线程 多目标 任务优先级 排序方法 演化算法 优化目标 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]