UML-based Cloud Application Modeling with Libraries, Profiles, and Templates

TLDR

Recent cloud modeling approaches have highlighted the need for cloud‑specific extensions to UML’s generic deployment language to capture diverse provider offerings. This paper introduces the Cloud Application Modeling Language (CAML) to express cloud‑based deployments directly in UML, facilitating migration to selected cloud environments. CAML is implemented as a UML internal language that employs a model library for deployment topologies and profiles that wire these topologies to cloud provider offerings. The authors show that UML templates can provide reusable deployment blueprints and establish conceptual mappings between CAML and the standardized TOSCA specification.

Abstract

Recently, several cloud modeling approaches have emerged. They address the diversity of cloud environments by introducing a considerable set of modeling concepts in terms of novel domain-specific languages. At the same time, general-purpose languages, such as UML, provide modeling concepts to represent software, platform and infrastructure artifacts from different viewpoints where the deployment view is of particular relevance for specifying the distribution of application components on the targeted cloud environments. However, the generic nature of UML’s deployment language calls for a cloud-specific extension to capture the plethora of cloud provider offerings at the modeling level. In this paper, we propose the Cloud Application Modeling Language (CAML) to facilitate expressing cloud-based deployments directly in UML, which is especially beneficial for migration scenarios where reverse-engineered UML models are tailored towards a selected cloud environment. We discuss CAML’s realization as a UML internal language that is based on a model library for expressing deployment topologies and a set of profiles for wiring them with cloud provider offerings. Finally, we report on the use of UML templates to contribute application deployments as reusable blueprints and identify conceptual mappings between CAML and the recently standardized TOSCA.