Results of behavioral synthesis, i.e. the architectures, are dependent on manifold constraints and therefore varying widely. Orinoco performs a timing or area driven scheduling and then computes upper and lower bounds of the power consumption for all possible allocation and binding solutions. This is done by mapping the allocation and binding problem to a graph covering problem and solving it using numerical techniques. Applying a set of heuristics Orinoco subsequently suggests a near to optimal architecture derived from the lower bound. Since the power consumption is dependent on the data processed, such data is taken into account during estimation and optimization.