Performance Analysis Method Study Of Loosely Coupled System Based On Aop

Loosely coupled systems are widely used in enterprise. Due to have dispersion and heterogeneous features, the performance of loosely coupled systems are difficult to be measured and analyzed. Traditional performance analysis methods of loosely coupled system concern on specific certain sub aspect applications, or costly because of needing to modify many points of the existing systems. The performance analysis method of dynamic and low cost for mature enterprise application systems will have important economic and social significance.Inversion of Control(IoC) mode is a design mode based on loosely coupled design thinking. It reduces and delays the dependencies of objects by transfers the dependencies to higher level framework. IoC realize the components'centrally deploy and test and improve the macro-system- intuitive structure and flexibility, which allows developers to focus on the business logic of the design and implementation. Aspect Oriented Programming(AOP) is one important step of programming ideological development history, is a useful complement to object-oriented programming. It is based on IoC mode. Through crosscutting the system into multiple aspects, we can dynamic weaving new components to the aspects by AOP framework's different enhance method, so finally we can into the objects inside.Based on IoC and AOP, this paper will present a new method to analyze the performance of mature enterprise loosely coupled systems. Using this method, we can analyze the performance of the enterprise system easily and costly. This paper will focus on how to subdivide the performance time information, how to use IoC container and AOP's proxy to monitor the performance information, how to compute the performance result from the performance information. Finally a experiment will be given to prove the performance analysis method's feasibility and accuracy.