Handling non-functional requirements for big data and IOT projects in Scrum

Abstract

Agile methodologies are gaining popularity at a lightning pace in the software industry. Prominent agile methodologies, such as Scrum and Extreme Programming (XP), help quickly deliver a system that meets Functional Requirements (FRs), while adapting to the changes in the customer requirements and feedback. However, Non-Functional Requirements (NFRs) by and large have been either mostly ignored or introduced and retrofit late in the software development cycle, oftentimes leading to project failures. With the recent, increasing shifts towards cloud and emphasis in big data in the software industry, NFRs, such as security and performance, have become more critical than ever before. This research proposes a novel approach to handling security and performance NFRs for projects involving big data and cloud, using Scrum. An industrial case study, conducted over a span of 9 months, shows that the approach helps deal with security and performance requirements individually as well as conflicts between them in an agile methodology.