A Novel Event Based Autonomic Design Pattern for Management of Webservices

• Published in: Advances in Computing and Information Technology pp. 142 - 151
• Main Authors: Mannava, Vishnuvardhan, Ramesh, T.
• Format: Book Chapter
• Language: English
• Published: Berlin, Heidelberg Springer Berlin Heidelberg 2011
• Series: Communications in Computer and Information Science
• Subjects: Adaptive Patterns; Autonomic computing; Case-based Reasoning; Dynamic service composition; Observer Pattern; Strategy Pattern; Stub; UDDI; Web Service
• ISBN: 3642225543; 9783642225543
• ISSN: 1865-0929; 1865-0937
• DOI: 10.1007/978-3-642-22555-0_16

A system is said to be adaptive if its behavior automatically changes according to its context.Systems based on the service-oriented architecture (SOA) paradigm must be able to bind arbitrary Web services at runtime.Web services composition has been an active research area over the last few years. However, the technology is still not mature yet and several research issues need to be addressed. In this paper, we propose an autonomic design pattern that describes the dynamic composition and adaptation of Web services based on the context. This pattern is primarily an extension of the Case-based Reasoning, Strategy, Observer Design Patterns.We proposed a framework where service context is configurable to accomodate the needs of different users and can adapt to dynamic changing environments.This permits reusability of a service in different contexts and achieves a level of adaptiveness and contextualization without recoding and recompiling of the overall composed services. The execution of adaptive composite service is provided by an observer model. Three core services, coordination service, context service, and event service, are implemented to automatically schedule and execute the component services, that adapt to user configured contexts and environment changes at run time. We demonstrate the benefits of our proposed design pattern by an experimental setup with implementation without generating stubs at the client side.