图类算法可重用设计及其实现

为了提高图算法生成效率和可靠性，该文提出一种将领域特征模型与构件组装技术相结合的可重用的图类算法开发方法.首先，通过对一族图算法的深入分析，揭示出图类算法领域的共性特征和可变特征，建立领域特征模型；然后，分析特征之间的交互过程，设计图类算法的可重用构件，并对构件依赖关系做出描述；最后，借助高可靠平台对算法构件进行开发，建立高可靠可重用构件库，进一步由构件组装出多种图算法，提高了图算法的开发效率和可靠性.实验表明开发出的图算法可重用构件库具有一定的实用性.

The Reusable Design and Implementation of Graph Algorithms Family

In order to improve the efficiency and reliability of graph algorithms generation,the reusable development method of graph algorithms that combines domain feature model and component assembly technology is proposed.First,a family of graph algorithms are analyzed to reveal the common features and variable features in the domain of graph algorithms and a domain feature model is established.Then,the interaction process between features is analyzed,the reusable components of graph algorithms are descigned and the component dependencies are described.Finally,the algorithm components with the help of PAR platform are developed,a library of highly reliable reusable components is established,and further the various graph algorithms are assembled by the components,the development efficiency and reliability of these graph algorithms are significantly improved.The experiments show that the developed reusable component library of graph algorithms has certain practicality.