Towards guidelines for building a business case and gathering evidence of software reference architectures in industry

Abstract

Background: Software reference architectures are becoming widely adopted by organizations that need to support the design and maintenance of software applications of a shared domain. For organizations that plan to adopt this architecture-centric approach, it becomes fundamental to know the return on investment and to understand how software reference architectures are designed, maintained, and used. Unfortunately, there is little evidence-based support to help organizations with these challenges. Methods: We have conducted action research in an industry-academia collaboration between the GESSI research group and everis, a multinational IT consulting firm based in Spain. Results:The results from such collaboration are being packaged in order to create guidelines that could be used in similar contexts as the one of everis. The main result of this paper is the construction of empirically-grounded guidelines that support organizations to decide on the adoption of software reference architectures and to gather evidence to improve RA-related practices. Conclusions: The created guidelines could be used by other organizations outside of our industry-academia collaboration. With this goal in mind, we describe the guidelines in detail for their use.

A Survey on the Benefits and Drawbacks of AUTOSAR

“…However, we remark that there exist organizations with similar contexts to AUTOSAR that could benefit from the results of this survey [18]. As Seddon et al suggests: "if the forces within an organization that drove observed behavior are likely to exist in other organizations, it is likely that those other organizations, too, will exhibit similar behavior" [21].…”

Benefits and drawbacks of software reference architectures: A case study

“…Also, vendors of the SRA are more likely to also be the provider that develops the applications, e.g., "once you define the SRA for a client organization, you have more options of developing applications on top of that SRA, since you know it" [H]. For this reason, the use of reference models is becoming popular among SRA vendors [9]. Table VIII.…”

“…We recognize that our results cannot be generalized to other organizations without further work. However, we claim that our study can serve as a basis for other similar organizations to study their SRA's benefits and drawbacks [9]. Other IT consulting firms that could be considered to some extent similar to everis are, for instance, Accenture [41] and Capgemini [42], as they use an industry-specific reference model to provide support to their clients to adopt SRAs, and they have similar professional roles to perform the associated tasks (see Fig.…”

“…The case study was conducted in two stages. First, the data related to SRA engineering at everis was collected and analyzed [9]. Second, this case study was designed to understand the benefits and drawbacks of nine SRA projects run at client organizations of everis.…”

Peer Review #2 of "Towards suitable description of reference architectures (v0.1)"

“…It aggregates knowledge about how to design software architectures of systems of a given application domain (Bass, 2013;Nakagawa et al, 2011). It includes domain business rules, standards and legislation, software and hardware elements, architectural styles and patterns, and best practices of software development in that domain, among other elements (Angelov et al, 2012;Martínez-Fernández et al, 2014;Nakagawa et al, 2011). Hence, the main purpose of reference architectures is to serve as a guide for the development, standardization, and evolution of systems (Nakagawa et al, 2014;Yimam and Fernandez, 2016;Martínez-Fernández et al, 2017).…”