Mapping the Emergence of the Field of DevOps in Management and Software Development
DevOps is an exciting new way of developing software that in recent years has emerged as a follow-up to agile software development. The movement focusses in particular on integrating the traditional silos of development and operations to enable continuous integration and delivery of software systems. It is becoming increasingly clear that the introduction of DevOps into an organisation is not only a technical challenge, but also requires an adjusted business model and hierarchy to be successful. It is the goal of this project to identify the outline of DevOps, as well as its technical and managerial implications for software development companies.
The project seeks to explore the influences and implications of DevOps adoption in both management and software development.
Our overall research objectives revolve around the empirical exploration of the adoption of a DevOps approach within organisations. Specifically, we are interested in: Why organisations adopt DevOps (what are the perceived performance or strategic benefits), how is it different to other approaches in software development, and what anticipated gains are acutally realised following adoption? How does DevOps adoption influence technical and social debt within software development processes? How is the management of software development processes influenced and affected by DevOps adoption and vice-versa? How do software development and IT operations roles, tasks, skills, tools and work identity change as DevOps is adopted?
In recent years, the DevOps literature has grown highlighting increasing academic and industrial research attention. As part of the project, we are currently undertaking a systematic review of this growing body of literature.
One thing that became clear from the early literature findings was that DevOps has proven problematic to define. There is a clear lack of rigour and methodological transparency as the bulk of definitions presented differed, and generally lacked any empirical value, with many of the cited sources no longer available. Following this, we have hosted an industrial focus group for DevOps practitioners in order to empirically evaluate and define DevOps. Two definitions were produced, which are currently being validated with the wider DevOps community through a survey. The analysis of the data is being undertaken in collaboration with industrial partners.
Our research activity also involves a six month diary study utilising the same tools participants use in their day to day roles. We are supporting the diary using a series of semi-structured interviews at key stages within the study period. This not only affords the option to probe diary data, but also acts as a control mechanism and potential alternative means of data collection. Our study focuses on software development and IT operations personnel within two UK based case study companies, currently adopting a DevOps approach. We will also apply the theoretical lens of job crafting to the collected qualitative data in order to explore the changes affecting individual job roles and work identity.
Collaboration: Norwich Business School
Stephen Jones, Fiona Lettice, Joost Noppen
Copyright© 2016, Norwich Business School and School of Computing Sciences, University of East Anglia, Norwich, Norfolk, United Kingdom. All Rights Reserved.