Given the complexity of biological systems, one may expect serious hurdles to encounter when it comes to designing optimization strategies for robust cellular performance. The metabolic flexibility and adaptive potential of prokaryotes is the mere cause of their successful applications, but these properties at the same time should be well understood to keep control over performance. This calls for a systems biological approach where all levels of cellular organization are integrated to describe quantitatively the input–output relationships of the organism as a whole. As exemplified above, genome-broad transcriptome analyses can be, and are, carried out with Synechocystis, and recently the first metabolomic data have been published (e.g. [46 ]). With the primary metabolic network being well resolved, with the data obtained by metabolic flux analysis [47,48] and with increased information on regulatory systems becoming available, in silico modeling of the organisms’ performance will be a strong tool soon. For example, a controlled flux distribution from GAP to biomass and the programmed product must be maintained in order to ensure a continuous process without the loss of viability. At the same time, maximization of production formation must be guaranteed. An example of the usefulness of metabolic flux balancing [49,50] to quantify flux distributions for a given condition has been recently provided by Fu [12 ] when simulating ethanol production by an engineered strain of Synechocystis.