Reference Architecture synthesis

Research Question

What are the benefits and drawbacks of reference architectures?

Search String

"software reference architecture"

Inclusion Criteria

• Any type of empirical study regarding software reference architecture benefits or drawbacks
• Any kind of outcome constructs used to observe software reference architecture is eligible for inclusion
• Forward and backward 'snowballing' should be considered for the included papers in order to look for additional studies to be included.

Exclusion Criteria

• Studies not reported in english
• The paper must be the primary source of the reported study or data. If the paper just reanalyze or review published study or data it should not be considered.

• The Concept of Reference Architectures
• Study Type : Observational
• Quality evaluation
• Evidence
• Constraints for the design of variability-intensive service-oriented reference architectures–An industrial case study
• Study Type : Observational
• Quality evaluation
• Evidence
• Software Reference Architectures - Exploring Their Usage and Design in Practice
• Study Type : Observational
• Quality evaluation
• Evidence
• Benefits and Drawbacks of Reference Architectures
• Study Type : Observational
• Quality evaluation
• Evidence
• A Survey on the Benefits and Drawbacks of AUTOSAR
• Study Type : Observational
• Quality evaluation
• Evidence

• all evidence v2
• Evidence based on the paper "A Survey on the Benefits and Drawbacks of AUTOSAR"
• Evidence based on the paper "Constraints for the design of variability-intensive service-oriented reference architectures–An industrial case study"
• Evidence based on the paper "The Concept of Reference Architectures"
• Evidence based on the paper "Benefits and Drawbacks of Reference Architectures"
• Evidence based on the paper "Software Reference Architectures - Exploring Their Usage and Design in Practice"
• Result
• all evidence
• Evidence based on the paper "A Survey on the Benefits and Drawbacks of AUTOSAR"
• Evidence based on the paper "The Concept of Reference Architectures"
• Evidence based on the paper "Constraints for the design of variability-intensive service-oriented reference architectures–An industrial case study"
• Evidence based on the paper "Benefits and Drawbacks of Reference Architectures"
• Evidence based on the paper "Software Reference Architectures - Exploring Their Usage and Design in Practice"
• Result

Research question

Seven effects caused by SRAs increased their belief values after the aggregation. These effects have greater confidence value that any effect before aggregation (i.e., greater confidence level than 40%), and have been reported by at least three out of the five studies

all evidence

all evidence v2

As a result of the aggregation process, several benefits of SRAs have considerably increased their belief value: interoperability of software systems, reduced developments costs, improved communication among stakeholders, reduced risk due to previous experiences, use of best practices, and reduced time-to-market. Regarding drawbacks, the studies agree that an SRAa SRA weakly negatively - negatively influence the required learning curve for developers before they could construct software systems.

The aggregation also identified three effects of SRAs that were contradictory in the literature: SRA complexity, flexibility of suppliers, and terminology conventions. We have analyzed the context of studies that reported these results, and elaborate hypothesis to explain these contradictions.