Information is everywhere, modern life depends on information. The faster we process data and generate meaningful outcomes, the better. Any IT project produces a lot of artifacts like software code, requirement documents, financial reports, traceability metrics, status reports, and so on. With all this documentation, we are often overloaded with information that makes the picture murkier. Having the wherewithal to capture the needed key performance indicators, there is still a gap between expectation and reality. This thesis is going to fill this gap between available metrics and desired indicators. The agile software development model has been around for a long time now. Today’s organizations are bigger than ever before and are competing in a changing landscape faster than ever before. Scaled agile models have helped the industry to ramp up on the human resources front but these models have not matched the needs of software maturity. Software status gathering and presenting tools are largely waterfall or small-scale agile-centered. The core of the agile approach is value creation but there is no single universal means to measure value. Our research intention is to present a standard value proposition in agile and provide metrics that are better suited for project managers. With help from empirical data, we not only identified the soar areas but also tried to remedy major concerns. To reach our goal, we used face-to-face interviews as our research strategy to get an account of firsthand experiences with the existing KPIs and what are the challenges. Our data collection was done through interviews with domain experts or people with extended knowledge of the subject. Finally, we performed a thematic analysis of the data to find what are the common themes emerging out of the shortcomings of the existing tools and what are the common desired requirements from KPIs. We have given a generic definition of value as well as proposed some KPIs that might come in handy for the practitioners of SAFe.

Software deployment is all of the activities that make a software system available for use. An examaple of such activites can be installations, updating and removing of software. Each deployment platform can differ in quality attributes such as performance, maintainability, etc. Docker, a deployment platform using container technology offering a more modular way of deployment has become increasingly popular over the past few years. The aim of the project is to find out how a deployment with Docker would affect the performance of a client­server application from the mining industry. Three scenarios were performed in a simulated environment for the existing as well as the potential deployment platform. Measures of round­trip time were made for both sets of the scenarios. Results show a general decrease in performance when running the application on Docker. This is especially seen in early stages of each scenario, where Docker in average has 15 times higher round­trip times than the existing platform. However, results gathered from each scenario suggests that while Docker does start much slower than the existing platform, in later stages of each scenarios, Docker manages to narrow the gap and be at most 1.14 times higher than the existing platform. In conclusion, while the deployment in which the existing platform is deployed on does out perform Docker, it still shows some promise and could potentially be a worthy option to look at for further work.