Architecture Assessment Model for System Evolution

Abstract

1Even though there has been some research on system evolution, there is no well defined vocabulary to indicate deviation of a system from desired goals. Further, there are no objective measures to indicate whether changes incorporated into a system as part of its evolution violates the integrity of the architectural design. Also, little research has been done to categorize the aspects of a software system that is subject to deviation as the system evolves.In this paper we develop a model for tracking software evolution and propose measures that will objectively indicate the extent of deviation or divergence in a software system. We also categorize the different aspects of software, changes to which can significantly impact usability as well as conceptual coherence.

Warnings: Violation symptoms indicating architecture erosion

“…Le et al [18,22] regarded architectural smells as structural symptoms and provided metrics to detect instances of architecture erosion by analyzing the detected smells. Bhattacharya et al [23] developed a model for tracking software evolution by measuring the loss of functionality (as evolution symptoms). Regarding the scope of our work, we focus on the nature of architecture erosion (i.e., violation symptoms) through code review comments in this work, which paves the way towards shedding light on architecture violations from the developers' perspective.…”

A Systematic Review on Architecting for Software Evolvability

“…business, technology, development process and organization. From business perspective, system requirements evolve because stakeholders' needs and expectations change, the context in which the software operate changes [9], and business models evolve [49]. From technology perspective, many unknown, uncontrollable technological and environmental constraints outweigh design principles [27].…”

On Creating Industry-Wide Reference Architectures

“…• Interface element dimension: operations, data and events. This is adapted from the abstract architectural model representation for service divergence evaluation, developed in [3]. That is, a divergence between two service interfaces can be further divided into operation divergence, data divergence, event divergence.…”