Extraction And Application Of Reusable Components In Legacy Systems

Over time,legacy systems can neither meet the new needs of users nor take advantage of new technologies,while maintenance costs continue to increase.In order to solve these problems,some users choose to redevelop applications by using new technologies,but the cost of this approach is too expensive.Correspondingly,it becomes a more reasonable choice to adjust and refactor legacy systems according to the application requirements and migrate them to the new operating platform.As a new emerging computing model,the influence of cloud computing in the computer field is gradually increasing.In order to make full use of the advantages of cloud computing platform such as fast speed and high efficiency,organizations need to make corresponding changes to the original software system according to the characteristics of the cloud computing model,and migrate the legacy system to the cloud platform.For partial cloud migration of legacy systems,the first problem to be solved is how to accurately extract reusable components in the system,and ensure reasonable component granularity and accurate and effective component description.Another core issue is how to find the reusable components that best meet the requirements of users in the process of component reuse.On the basis of using clustering algorithm to extract reusable components in literature [1],this paper proposes a new method of component extraction for systems using common development frameworks,and combines it with method of clustering extraction effectively to form a new strategy for component extraction.At the same time,this paper improves the matching method of target service and candidate components in literature [2],and uses artificial intelligence algorithm to achieve text matching,which helps users to find components that meet requirements more accurately.Based on the above research,support tools were developed to realize automatic component extraction and service matching,and experimental verification was carried out at the same time.The experimental results show that for the system using the development framework,the tool has better efficiency of component extraction,the results of component extraction are more in line with the actual requirements,and the matching process with the target service is more accurate.Related work has a positive effect on the reuse of components of the business logic in the migration process.