软件定义网络系统的可靠性研究
2017-09-08夏飞
夏飞



摘 要: 可靠性是评价软件定义网络(SDN)系统的一项关键指标,为了提高SDN系统的整体可靠性,针对系统的软件可靠性进行研究,提出一种基于量子群优化算法的SDN系统软件可靠性评估模型。首先对SDN系统的软件可靠性研究现状进行分析,并建立相应的数学模型,然后采用量子粒子群算法对数学模型进行求解,并通过具体数据对软件可靠性评估模型的性能进行仿真测试。结果表明,量子粒子群算法可以对SDN系统的软件可靠性进行准确评估,很好地描述了SDN系统的软件可靠性变化趋势,而且求解结果要优于其他模型,实际应用价值更高。
关键词: 量子粒子群算法; 软件定义网络; 软件可靠性; 数学模型
中图分类号: TN915.08?34 文献标识码: A 文章编号: 1004?373X(2017)17?0020?03
Research on reliability of software defined network system
XIA Fei
(Information Communication Company, State Grid Jiangsu Electric Power Company, Nanjing 210024, China)
Abstract: The reliability is a key index to evaluate the software defined network (SDN) system. In order to improve the overall reliability of SDN system, the software reliability of the system is studied, and then a SDN system′s software reliability evaluation model based on quantum?behaved particle swarm optimization algorithm is proposed. The research statue of software reliability of SDN system is analyzed. The corresponding mathematical model was established, and solved with the quantum?behaved particle swarm optimization algorithm. The simulation test was performed for the performance of the software reliability evaluation model by means of specific data. The results show that the quantum?behaved particle swarm optimization algorithm can evaluate the software reliability of the SDN system accurately, describe the software reliability variation trend of SDN system perfectly, and its solving result is better than that of other models and has high practical application value.
Keywords: quantum?behaved particle swarm algorithm; software defined network; software reliability; mathematical model
0 引 言
随着网络技术应用的不断深入,网络用户数量迅速增加,为了满足应用要求,产生了一种新型网络架构——软件定义网络(SDN)系统[1?3]。无论哪一种网络系统,可靠性是一项关键的性能指标。SDN系统的可靠性包括硬件可靠性和软件可靠性两个研究方向,本文主要针对软件可靠性进行分析[4]。
为了提高SDN系统的可靠性,针对系统中的软件可靠性进行研究。提出一种量子粒子群算法的SDN系统的软件可靠性评估模型,仿真测试结果表明,量子粒子群算法可以对SDN系统的软件可靠性进行准确评估,十分准确地描述了SDN系统的软件可靠性变化趋势。
1 软件可靠性评估的相关研究
SDN系统在工作过程中,有时会出现故障,故障一旦被消除,那么SDN的软件失效率下降,因此失效率是评价SDN系统软件可靠性的一个重要指标[5]。……
