DSD focuses on responsiveness – the probability of successful operation within a deadline even in the presence of faults – in modern service networks, more specifically, the responsiveness of the service discovery layer, which is used to publish and discover instances providing a specific service type. To date, dependability assessment of services almost exclusively neglects discovery, albeit being a critical part of successful service usage. In future service network scenarios with an abundance of services ranging from minimal to highly complex, composed functionality, the importance of an automated and robust discovery is expected to increase significantly. The goals of this project are to (1) develop and validate a stochastic model for responsiveness assessment based on data solely collected by existing mechanisms within the SD protocols. (2) Propose an adaptive discovery system for responsiveness optimization. (3) Develop and validate a model to predict network breakdowns based on discovery response times.