Browsing by Author "Jokinen, Oskari"

The objective of this study is to get insight on the usage of automated build and deployment pipelines by software development teams at Avaintec Oy in the context of DevOps methodologies. Automated deployments are used widely in today’s software development and they allow for complex installation operations and testing to be done in a time-saving manner as less manual work is required. While there is a lot of research in the area of DevOps and automation pipelines, standardised ways of working and best practices are still vague in many ways. Studying the effects of DevOps and the use of the automation pipelines in the scope of the development team is important to establish practices and ways of working that can best support the work of the developers and also be cost effective for the companies. The research methods of this study are a literature review of the most recent literature on using automation in software development, and a case study of the process of taking DevOps methodologies and a Continuous Deployment pipeline into use at Avaintec Oy. The literature is analysed with the help of a theme matrix. The case study was conducted by interviewing selected members of development teams at Avaintec. The interviews were recorded and transcribed, and then analysed with a theme matrix. The output from the literature review is reflected to the output from the interviews in the case study section. The conclusions from this study indicate that development teams find DevOps and DevOps pipelines useful, despite the learning curve in the new tools and methodologies and the amount of initial setup work. The evolution of an automated build and deployment pipeline should be continuous and pipelines should be built incrementally, both of which were true for the studied case. The engagement of developers to the Operations side is a challenge present in both literature and the studied case, as developers at times are not that familiar with the pipeline. Apart from the pipeline, products should also be developed to be suitable for DevOps and automatic deployments, microservices seem to provide advantage in this.