摘要
RESTful服务架构风格的Web服务比SOAP或XML-RPC更加简洁,已被广泛的应用在企业的系统服务中。如何有效地保障其服务质量和安全可靠变得尤为重要。软件测试是保障软件架构服务正常的有效手段,为了满足日益发展的软件技术对可靠性保障的需求,针对软件可靠性的测试研究也越来越重要。针对RESTful资源服务的可靠性进行探究,提出基于马尔可夫链的使用模型,在采用重要抽样方法保证可靠性无偏估计的情况下,通过启发式遗传算法种群迭代选择过程,调整优化使用模型的操作状态转移概率,增加关键操作的执行次数,解决软件统计测试的时间和开销费用较大的难题。通过实验仿真结果,表明与已有的模拟退火方法和标准统计测试相比,该方法能够快速地生成最优矩阵,得到失效概率更小的方差,提高软件统计测试的效率,达到测试的充分性。
A restful service architecture style web service is more concise than SOAP or XML-RPC, and has been widely used in enterprise systemservices. How to effectively ensure the quality of service and safety becomes more and more important. Software testing is an effective meth-od for normal security software architecture service, in order to meet the growing demands for reliability, security, software technology studyof software reliability testing is becoming more and more important. For restful resource service reliability, puts forward the use modelbased on the Markov chain, in the importance sampling method is adopted to ensure reliability under the condition of unbiased estimation,through iterative heuristic genetic algorithm population selection process, adjust the operating model of optimizing use of state transitionprobability, increase the number of executions key operations, to solve the software statistical test time and the problem of overhead expens-es is larger. Finally, the simulation results, through the experiment shows that with the simulated annealing method compared with the stan-dard statistical tests, the method can quickly generate the optimal matrix, get the failure probability of the smaller variance, improve the ef-ficiency of the software of statistical test, achieve the adequacy of the test.
作者
江金莲
关婷婷
JIANG Jin-lian;GUAN Ting-ting(College of Information Engineering,Shanghai Maritime University,Shanghai 201804)
