ASCETiC: Adapting Service lifeCycle towards EfficienT Clouds
ASCETiC addresses the issue of energy efficient computing, specifically in the design, construction, deployment and operation of Cloud services. Its approach focuses firstly on the identification of the missing functionalities to support energy efficiency across all cloud layers, and secondly on the definition and integration of explicit measures of energy requirements into the design and development process for software to be executed on a Cloud platform. Therefore, in this context, there are a number topics that can be considered in future work programmes:
- Energy efficiency and big data: by reduction of data movement, by eliminating redundancies without affecting fault tolerance, and by re-using big data analysis results.
- Cloud sustainability: investigation on heterogeneous hardware usage in clouds and new virtualisation means (aka containers) to improve Cloud sustainability, as well as extending sustainability aspects to consider network operation energy consumption.
ASCETiC important results so far include:
- Year 1 architecture reference implementation prototype: this milestone concentrates on delivering energy awareness in all cloud system components. Monitoring and metrics information can be measured at IaaS level and propagated through the other layers of the Cloud stack (PaaS, SaaS) considering static energy profiles.
- Early evaluation of ASCETiC cloud stack focusing on software re-engineering by means of the SaaS layer modelling tools. For example, the use case News as an Asset (NewsAsset), which is an end-to end multimedia cross-channel solution for evolving News Agencies, Broadcasters and Publishers, is able to express energy monitoring requirements with augmented UML in diagrams from process or logical view.
Year 1 ASCETiC architecture reference implementation is expected to be launched in November 2014.
The project has already identified a number of toolkit components that will lead to direct exploitation, e.g. SaaS modelling tool, energy modeller, Virtual Machine Contextualizer. The exploitation of the entire ASCETiC cloud architecture and its reference implementation (SaaS, PaaS, IaaS) is envisaged through the creation of a spin-off company by the end of the project.
Some collaboration has started with Eco2Clouds, which addresses the ecological implications of Cloud infrastructures. Communications are taking place to jointly investigate strategies that can ensure effective application deployment and execution on Cloud infrastructures while aiming to reducing energy consumption. More specifically, how Eco-metrics collection, cloud application deployment, and evaluation mechanisms/optimisation algorithms are considered in both projects. Early contacts have taken place with CloudScale whose aim is to make cloud systems scalable by design so that they can exploit the elasticity of the cloud. This is of interest to ASCETiC as it aims to enable the analysis of software in terms of energy efficiency at the various stages of the cloud service lifecycle: construction, deployment, and operation.