Runtime Integration and Testing for Highly Dynamic Service Oriented ICT Solutions -- An Industry Challenges Report

Abstract

Modern Information and Communications Technology (ICT) solutions are often widely distributed and highly dynamic service oriented architectures (SOA) with stringent availability requirements. Availability implies that SOA must be reconfigured, updated and maintained during runtime, while retaining their overall operational integrity. This requires that much of the adaptation, integration, configuration and testing activities typically performed offline, during development time, now have to be done online, during runtime. Current component-based runtime platforms such as SOA realize the technological foundations for runtime reconfiguration and maintenance. However, because software engineering methodology has not kept pace with the rapid leap forward in platform technology, adequate methods, techniques and tools for dealing with runtime integration and testing are not yet available. This paper discusses the industry challenges and open issues of integrating and testing SOA infrastructures during runtime.

An extended framework for supply chain risk management: incorporating the complexities of emerging industries and large-scale systems

“…Traditional industries continue to have SCR-related challenges in testing of components (Gilbert and Wisner, 2010), software (Bertolino, 2007), communications technologies (Greiler et al, 2009), and in large engineered systems such as space flight systems (Miller et al, 2008) and nuclear reactor materials (Yvon and Carré, 2009). …”

Adopting and Evaluating Service Oriented Architecture in Industry

“…We are currently involved in an ongoing research project that investigates the primary industry challenges of runtime integration and testing. We believe that that runtime evolution and testing of SOA can only be implemented successfully if, especially, the communication between stakeholders, and the test-isolation and test-awareness of services are improved [32].…”

Spectrum-based fault diagnosis for service-oriented software systems

“…These benefits are mainly attributable to the loose coupling of services, facilitated through modern underlying communication platforms, and their natural disposition to dynamic deployment, reconfiguration, and evolution. However, this nature of service-oriented system also presents many challenges [1], particularly concerning quality assurance. The fact that service-based applications only fully materialize when deployed in production, i.e., ultra-late binding [2], renders many of the traditional (offline) quality assurance methods useless [3].…”