The rising energy consumption of the ICT industry has triggered a quest for more green, energy efficient ICT solutions. The role of software as the true consumer of power and its potential contribution to reach sustainability goals has increasingly been acknowledged. At the same time, it is shown to be a complex problem to address. This research focuses on enabling software producing organizations, e.g. independent software vendors and open source foundations, to deliver green software products and thereby contribute towards fulfilling climate goals.
The dissertation is focused on the following main research question:
How can software producing organizations be in control of the energy consumption of their software products?
To provide an answer on this question, the dissertation is divided in three parts totaling six chapters. The first part of the dissertation is concerned with the ability to measure the energy consumption of software products, which is a prerequisite for the research. The second part relates energy consumption to software architecture, to enable the analysis of energy efficacy on architectural level. The last part studies energy consumption, and its consequences and effects in an industrial context, to create awareness on the topic. Overall the dissertation follows a design science research methodology with the objective to integrate energy consumption in software engineering. The activities comprising the methodology were also performed separately for each research endeavor presented in the individual chapters. Through multiple experiments and case studies, some using commercial software products, the required artifacts were created, their application demonstrated and their implications on the software engineering process evaluated. Combined, the artifacts provide a software producing organization the knowledge and tools to be in control of the energy consumption of their software products throughout the complete software lifecycle.