Adaptive SOA stack and discovery framework for redundant services

Abstract

In recent years marketplace agility and dynamic business needs require organizations to respond rapidly and cost-effectively to business demands. Service Oriented Architecture (SOA) is one of the most successful ways to design and integrate distributed systems. SOA has evolved to adopt vision of Service Oriented Enterprise. In order to realize this vision it is important to align IT with business demands. Agility and services driven business models have put demands on quality aware and adaptive SOA systems. With enterprises formally moving towards adaptive processes and systems there is a need to execute and implement Adaptive SOA. This adaptation should support autonomous behavior of distributed services at run-time. Henceforth SOA implementation stack also need to evolve and exhibit adaptive behavior. With adoption of SOA becoming a norm it is common to have multiple services providing same functionality. Discovery of appropriate service will be based on consumers requirement of different Quality of Service parameters offered by these services. This study introduces an adaptive framework for discovering redundant services leveraging topological metadata. This framework re-routes consumers to an appropriate service in case of failure or performance degradation. Results are demonstrated on implementing the framework for calculating round trip time for co-located and remote clients. Results indicate advantage of adopting a topologically aware redundancy management adaptive framework.