Model-Driven Engineering for Analysing, Modelling and Comparing Cloud Computing Service Level Agreements

In cloud computing, service level agreements (SLAs) are critical, and underpin a pay- per-consumption business model. Different cloud providers offer different services (of different qualities) on demand, and their pre-defined SLAs are used both to advertise services and to define contracts with consumers. However, different providers express their SLAs using their own vocabularies, typically defined in terms of their own tech- nologies and unique selling points. This can make it difficult for consumers to compare cloud SLAs systematically and precisely. We propose a modelling framework that pro- vides mechanisms that can be used systematically and semi-automatically to model and compare cloud SLAs and consumer requirements. Using MDE principles and tools, we propose a metamodel for cloud provider SLAs and cloud consumer requirements, and thereafter demonstrate how to use model comparison technology for automating differ- ent matching processes, thus helping consumers to choose between different providers. We also demonstrate how the matching process can be interactively configured to take into account consumer preferences, via weighting models. The resulting framework can thus be used to better automate high-level consumer interactions with disparate cloud computing technology platforms.