An Approach to Evolving Legacy Software System into Cloud Computing Environment

An Approach to Evolving Legacy Software System into Cloud Computing Environment

Shang Zheng (De Montfort University, UK), Feng Chen (De Montfort University, UK), Hongji Yang (De Montfort University, UK) and Jianzhi Li (De Montfort University, UK)
DOI: 10.4018/978-1-4666-2854-0.ch009


Cloud computing is a new paradigm for the intent of distributed resources sharing and coordinated problem solution. Affected by the Cloud trend and Service-Oriented need, many existing software systems will become legacy systems. These legacy software systems will need Cloud Oriented reengineering, which can facilitate the legacy systems reusable in Cloud Oriented architecture and allow the integration of legacy resources with Cloud features. This research focuses on establishing a general framework to assist with the evolution of legacy systems into Cloud environments. The methodology includes various phases, which use reverse engineering techniques to comprehend and decompose legacy systems, represent legacy resources by XML as Cloud component and integrate these Cloud components into Cloud environment. In this research, a legacy banking system has been chosen as a case study to prove the feasibility of the proposed approach. The legacy banking system can be transformed to run as a Service-Oriented Cloud application, which illustrates the proposed approach is powerful for utilising reusable legacy resources into Cloud environment.
Chapter Preview

2.1. Resource Oriented Software Evolution

In (Huang et al. 2003), various issues and challenges in web-based systems development and maintenance are described, especially in the public domain. Evolvable web-based system architecture is introduced for management systems, and some designs and implementation tradeoffs are discussed as well. (Stroulia et al. 2000) suggest applying conventional reverse engineering techniques, such as code analysis and clone detection, to reduce duplicated content and to maintain web-based systems. The developed system includes the HTML parser and analysers that separate content from layout by integrating into HTML pages scripts for retrieving the dynamic data from special database.

In (Tonella et al. 2003), the researchers propose a cluster method, which is based on the automatic extraction by the keywords in web pages. Lixto (Estievenart et al. 2003) is a wrapper generation tool that is well suitable for building HTML/XML wrappers. (Moreira et al. 2001) propose an approach to integrating WWW information, which is based on the development of a canonical domain model in XML and the wrapping of existing WWW applications with wrappers capable of communicating about entities in this common model with the applications and with an intermediary mediator.

Complete Chapter List

Search this Book: