An Architectural Approach for Self-Managing Security Services

Self-managing systems are the focus of an increasing research activity since can provide a more robust way of building software artifacts. In this paper, we focus on self-managing systems for adapting the security concern. Providing security as a one-size-fits-all solution results in a system that is far too rigid to accommodate the needs of different application domains. Security mechanisms can depend on the hardware that is available and can be changed over time. Offering a system that can cope with such change without the need of major re-engineering is a major challenge. Moreover, not at all time the same security level has to be maintained. A system that can adapt its security mechanisms to the actual threat level can deliver better performance. In this paper, we provide an architecture for self-adapting the security concern to the actual context. Context is provided in terms of trust and threat values. Moreover, contextual information on resource consumption (e.g., memory and bandwidth) and system status (e.g., availability) can be provided.