Logo des Repositoriums

Auflistung nach Autor:in "Wilke, Claas"

1 - 3 von 3
  • Konferenzbeitrag
    Energy labels for mobile applications
    (INFORMATIK 2012, 2012) Wilke, Claas; Richly, Sebastian; Püschel, Georg; Piechnick, Christian; Götz, Sebastian; Aßmann, Uwe
    In recent years the usage of mobile devices and the expansion of their functionality by installing further applications have become very popular. Their frequent usage causes much faster battery discharging, and thus drastically limits the uptime of the devices and their applications. Hence, investigating and reducing the power consumption of mobile applications is one of the current, central challenges in software engineering. In this paper we propose an approach for profiling the power consumption of mobile applications and comparing their consumption for similar services. We show by example that such differences can be identified for two well-known email clients. We envision a repository or market place that allows users comparing and selecting applications based on energy labels and their personal requirements.
  • Konferenzbeitrag
    Energy profiling as a service
    (INFORMATIK 2013 – Informatik angepasst an Mensch, Organisation und Umwelt, 2013) Wilke, Claas; Richly, Sebastian; Götz, Sebastian; Aßmann, Uwe
    Energy optimization of software systems has emerged as one of the biggest challenges for the development of future software applications. Especially the efficiency of mobile applications has become a major driver, as limited battery capacities force applications to cause as few energy consumption as possible to increase the devices' uptimes. Although many research groups have investigated approaches to energy optimization of mobile applications, still, each group is building and using its own methodology to evaluate the impact of their optimizations on the devices' energy consumption. This variety of different evaluation methodologies hinders the reproducibility of individual research group's findings as well as their comparability with similar research results. In this paper, we propose profiling as a service by exposing a hardware-based energy profiling infrastructure as a web service reusable for other research groups and application developers. We discuss possible use cases for our profiling service as well as limitations. Besides, we present its transparent integration into software development tools such as Eclipse.
  • Zeitschriftenartikel
    Towards Predictive Self-optimization by Situation Recognition
    (Softwaretechnik-Trends: Vol. 33, No. 2, 2013) Götz, Sebastian; Schöne, René; Wilke, Claas; Mendez, Julian; Aßmann, Uwe
    Energy efficiency of software is an increasingly important topic. To achieve energy efficiency, a system should automatically optimize itself to provide the best possible utility to the user for the least possible cost in terms of energy consumption. To reach this goal, the system has to continuously decide whether and how to adapt itself, which takes time and consumes energy by itself. During this time, the system could be in an inefficient state and waste energy. We envision the application of predictive situation recognition to initiate decision making before it is actually needed. Thus, the time of the system being in an inefficient state is reduced, leading to a more energyefficient reconfiguration.