Towards an extended misuse case framework for elicitation of cloud dependability requirements

Abstract

Cloud computing has emerged as a foremost and influential technology driver of business initiatives in recent times. Although, security is the most discussed concern on cloud platforms, other dependability attributes such as reliability, availability, robustness and the likes are also important for successful cloud implementations. So far, existing approaches of dependability requirements elicitation have focused on eliciting individual dependability attributes separately even though dependability attributes are known to be interrelated and interdependent in terms of their potential impact on developed systems. Thus, it is difficult to gain a prompt and holistic understanding of the dependability constraints that a system must satisfy, which has limited the efficiency and effectiveness of the dependability requirements engineering (DRE) process. There is a dearth of approaches that can support an integrated view of a system's dependability attributes under a single requirements elicitation framework. In this paper, we propose new visual notational extensions to the existing misuse case (MUC) model in order to compose an integrated framework that can support the elicitation of various cloud dependability requirements. To do this, we conducted an analytical evaluation of some proposed notational extensions based on the nine principles of constructing visual notations in software engineering as proposed by Moody. Also, by using a case study from the mobile banking domain, a trial experiment was undertaken to assess the capability of the extended MUC framework to support the elicitation of multiple cloud dependability requirements. The result of the pilot experiment shows that the extended misuse case driven technique is credible and viable for the elicitation of cloud dependability requirements.