An Energy-Aware Brokering Algorithm To Improve Sustainability In Community Cloud
Title | An Energy-Aware Brokering Algorithm To Improve Sustainability In Community Cloud |
Publication Type | Conference Proceedings |
Year of Conference | 2017 |
Authors | Giacobbe, M., M. Scarpa, R. Di Pietro, and A. Puliafito |
Conference Name | Proceedings of the 6th International Conference on Smart Cities and Green ICT Systems (SMARTGREENS 2017) |
Volume | 1 |
Pagination | 166-173 |
Date Published | 04/2017 |
Publisher | Institute for Systems and Technologies of Information, Control and Communication (INSTICC) |
Conference Location | Porto, Portugal |
ISBN | 978-989-758-241-7 |
Keywords | cloud computing, Community Cloud, Energy-aware Brokering, Green Cloud, Low Carbon, Resource allocation, Sustainability, Virtualization |
Abstract | Cloud computing is a paradigm for large scale distributed infrastructures, platforms or software services which represents a hot topic in Information Technology (IT) recently in both industrial and academic areas. Its use is motivated by the possibility to promote a new economy of scale in different contexts. Along with the well-known public, private and hybrid Cloud models, the Community Cloud is an emerging concept based on a deployment model in which a Cloud infrastructure allows a specific community of consumers to share interests, goals and responsibilities. It can be owned and managed by the community, by a third party, or a combination of them. In such scenario, new low-carbon strategies at Cloud sites are necessary to allow those latter to reduce the consumption in presence of a massive exploitation of IT services. Therefore, balancing performances with both sustainability and cost saving concepts is a challenge. In this paper, we present a low carbon strategy designed to make the best choice in resources allocation, based on sustainability, availability and costs. The proposed energy-aware Brokering Algorithm (eBA) allows to push down carbon dioxide emissions through the Community Cloud ecosystem, by running instances at the most convenient sites. |
URL | http://www.scitepress.org/DigitalLibrary/PublicationsDetail.aspx?ID=8XfUdAkq8Pk=&t=1 |
DOI | 10.5220/0006300201660173 |