Formalizing self-adaptive clouds with knowlang

Abstract

Cloud computing emerged as a paradigm offering new benefits to both social networking and IT business. However, to keep up with the increasing workload demand and to ensure that their services will be provided in a fail-safe manner and under consideration of their service-level agreement, contemporary cloud platforms need to be autonomous and self-adaptive. The development of self-adaptive clouds is a very challenging task, which is mainly due to their nondeterministic behavior, driven by service-level objectives that must be achieved despite the dynamic changes in the cloud environment. This paper presents a formal approach to modeling self-adaptive behavior for clouds. The approach relies on the KnowLang language, a formal language dedicated to knowledge representation for self-adaptive systems. A case study is presented to demonstrate the formalization of Science Clouds, a special class of self-adaptive clouds providing a cloud-scientific platform.