Energy-aware Data Stream Management

With the increasing use of data stream management systems (DSMS's) in energy-critical areas the energy consumption of the DSMS itself is gaining growing importance. However, processing data streams in an energy-aware manner is still an open research question. In this paper, we propose a very first concept for a DSMS architecture which treats the energy consumption as a first-class-citizen. Therefore, we rely on data synopses and anytime algorithms in order to minimize the data quantity and the number of iteration which minimizes the overall energy consumption.