Towards Service Co-evolution in SOA Environments: A Survey

Tran, Huu Tam; Nguyen, Van Tu; Vinh, Phan Công

doi:10.1007/978-3-030-67101-3_19

Cited by 2 publications

(4 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Service evolution is an active field of research for more than 15 years. For a recent survey of previous efforts in this field, the interested reader can refer to the work of Tran et al 4 Following, we discuss our contribution with respect to the state of the art on service evolution. We organize the discussion in three parts, based on our research questions and the overall research goal of our study.…”

Section: Related Workmentioning

confidence: 99%

“…Service changes can take place at different levels of abstraction. In particular, Tran et al 4 survey previous research efforts that focus on interface, semantic, protocol and process changes. Similar classifications of the different kinds of changes involved in service evolution are provided by Treiber et al 18 and Wang and Wang 19 .…”

Section: Related Workmentioning

confidence: 99%

“…The co‐evolution of service orchestrations/choreographies with the orchestrated/choreographed services 4 is another interesting research issue that involves several different problems. For instance, Baresi et al 13 proposed an approach for the consistent dynamic evolution of service choreographies.…”

Section: Related Workmentioning

confidence: 99%

“…In general, service evolution is a challenging research issue with several interesting approaches that emerged so far, towards dealing with this issue. In particular, the state of the art on service evolution includes three central topics, specifically: (i) change‐detection tools , that is, tools that enable the detection of changes in subsequent service releases, and empirical studies that employ such tools to investigate the evolution properties of real‐world services; 4–8 (ii) frameworks that facilitate the management of service changes, both for the service providers, and the developers of the systems that depend on the evolving services; 4,9–14 and, (iii) patterns that specify best practices allowing to tackle the evolution of services in a more systematic and easy way 15,16 …”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

The athletic heart syndrome in web service evolution

Zarras

Dinos

Vassiliadis

2021

J Software Evolu Process

View full text Add to dashboard Cite

Despite the particular standards, technologies, and trends (W3C, RESTful, microservices, etc.) that a team decides to follow for the development of a service‐oriented system, most likely the team members will have to use one or more services that solve general‐purpose problems like cloud computing, networking and content delivery, storage and database, management and governance, and application integration. Typically, general‐purpose services are long‐lived, they have several responsibilities, their interfaces are complex, and they grow over time. The way that these services evolve also affects the evolution of any system that will depend on them. Consequently, the selection of the particular services that will be used is a main concern for the team. In this paper, we report a pattern, called the athletic heart syndrome, which facilitates the selection of services that evolve properly. Patterns specify best practices that emerge from multiple real‐world cases. In our context, the athletic heart syndrome comes out from a study that concerns the evolution of a set of popular, long‐lived Amazon services that cover different domains. According to the athletic heart syndrome, the developers should select services whose heartbeat of changes looks like the heartbeat of an athlete when he is at rest. Specifically, the heartbeat of changes should consist mostly of calm periods, interrupted by few spikes of change. Similarly, the incremental growth of the services should involve mainly calm periods of maintenance, separated by spikes of growth. Selecting services that adhere to the pattern signifies high chances that the services evolve to deal with changing requirements. The pattern further guarantees that the service evolution involves both the expansion of the services with new functionalities and the maintenance of existing ones. The pattern also assures that the complexity increase in the service interfaces will be smooth and tolerable. Finally, conformance with the pattern implies that the growth of the services will be predictable.

show abstract

Section: Related Workmentioning

confidence: 99%