Abstract

Flood water simulations have to be accurate in order to prevent losses and must therefore be based on well calibrated rainfall runoff models of sufficient complexity. IaaS cloud computing has be proven useful as a fast and cost-efficient extension of local computers, in the case, using the calibration of the hydrologic modeling system ArcEGMO as an example. On the other hand, cloud-based systems cannot serve as a replacement for a local computer, as long-term usage is still too expensive and there surely are performance differences regarding the offered architectures. Ultimately, the usage of cloud computing as a viable extension of local computing capacities is only reasonable if state of the art computer technologies with a fair pricing sytem can be offered by IaaS providers.