Service description languages in cloud computing: State-of-the-art and research issues
Service Oriented Computing and Applications
School of Science
The continuous growth of cloud computing environment is supported by the automated provisioning of cloud services, which allows cloud users to dynamically procure and deploy their required services over the internet. However, to describe key characteristics of a cloud service, each cloud service provider uses its own service description language (SDL) utilizing its underlying syntax and semantics coupled with models and standards to fulfill its own objectives (such as deployment, provisioning, modeling, discovery and composition). This prevents cloud computing community to adopt and enforce a standard mechanism for SDLs that limits the ability of automation of cloud services and results in vendor lock-in problem for cloud users. In this paper, we investigate different techniques of cloud SDLs by focusing on their main purpose of use in different cloud service operations including deployment and provisioning, modeling and composition, discovery and selection, and service level agreement. We use a common comparison criteria to classify existing literature on cloud SDLs with the goal to identify their common features in different service operations. Based on identified gaps in the literature, research issues have been structured to develop the foundations of a standard cloud SDL in order to develop high-quality cloud applications of the future.