Model-Driven Elasticity Management with OCCI

Abstract

Elasticity is considered as a fundamental feature of cloud computing where the system capacity can adjust to the current application workloads by provisioning or de-provisioning computing resources automatically and timely. Many studies have been already conducted to elasticity management systems, however, almost all lack to offer a complete modular solution. In this article, we propose <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">MoDEMO</small> , a new elasticity management system powering both vertical and horizontal elasticities, both VM and Container virtualization technologies, multiple cloud providers simultaneously, and various elasticity policies. <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">MoDEMO</small> is characterized by the following features: it represents ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i</i> ) the first system that manages elasticity using Open Cloud Computing Interface (OCCI) model with respect to the OCCI standard specifications, ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ii</i> ) the first unified system which combines the functionalities of the worldwide cloud providers: Amazon Web Services (AWS), Microsoft Azure and Google Cloud Platform (GCP), and ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">iii</i> ) allows a dynamic configuration at runtime during the execution of the application. <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">MoDEMO</small> permits to timely adapt resource capacity according to the workload intensity and increase application performance without introducing a significant overhead.