Research On Software Defect Diagnosis Of SOA Based On Spiking Neural P Systems

Software defect is the deviation which existing in software and can cause software defect and damage the security of software under a certain condition.In the traditional software system,the software with high coupling,high in reliability and scalability and so on.so when a module in the system error will cause a great damage to system,therefore,in order to solve the faults of software and improve software reliability and scalability,is the key point of the currently research.Software defect diagnosis is a problem of binary classification,which divided software into defective part and no defect part,and then analyze the defect parts,and make a defect prediction to a new software module.In order to solve these problems,the service-oriented architecture software system has been put forward,which is a component model and associate the different modules with the neutral interface,which is independent of the hardware platform,operating system and programming language,so a service-oriented software architecture can be applied to other multiple applications and business processes,and the binding and service between service requester and the service provider should be loose.Therefore,it is of great significance to study the diagnosis and treatment of service composition.Spiking neural system(SN P systems)is a neural like Computing Models which is similar to network structure of neural system.Due to its strong learning ability,it can also be used to solve many practical problems.In this paper,It is the first time that the spiking neural system has been applied to solve the problem of service composition diagnosis.The main research contents are as follows:Firstly,the definition of the model of service composition is given by the spiking neural p system with colored spike.Then,using the spiking neural P system with colored spike to model the available services,components and connectors in the service composition.After that,the system can deal with the failure in any case is analyzed and then the validity of the model has been proved through theoretical analysis,at the same time,compare the proposed method with the method of Petri net through experiment,the results showed that the efficiency and stabilityof the method with SNP system with colored spike has a certain advantagein the fault handling,Which can timely deal with the service composition failure and make the system back to normal quickly.