Facial reconstruction

Search LJMU Research Online

Browse Repository | Browse E-Theses

Cloud-SEnergy: A Bin-Packing Based Multi-Cloud Service Broker for Energy Efficient Composition and Execution of Data-intensive Applications

Baker, T, Aldawsari, B, Asim, M, Tawfik, H, Maamar, Z and Buyya, R (2018) Cloud-SEnergy: A Bin-Packing Based Multi-Cloud Service Broker for Energy Efficient Composition and Execution of Data-intensive Applications. Sustainable Computing : informatics and systems. ISSN 2210-5379

[img] Text
R1-final-NoTrackChange.pdf - Accepted Version
Restricted to Repository staff only until 26 May 2020.
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (1MB)

Abstract

The over-reliance of today's world on Information and Communication Technologies (ICT) has led to an exponential increase in data production, network traffic, and energy consumption. To mitigate the ecological impact of this increase on the environment, a major challenge that this paper tackles is how to best select the most energy efficient services from cross-continental competing cloud-based datacenters. This selection is addressed by our Cloud-SEnergy, a system that uses a bin-packing technique to generate the most efficient service composition plans. Experiments were conducted to compare Cloud-SEnergy's efficiency with 5 established techniques in multi-cloud environments (All clouds, Base cloud, Smart cloud, COM2, and DC-Cloud). The results gained from the experiments demonstrate a superior performance of Cloud-SEnergy which ranged from an average energy consumption reduction of 4.3% when compared to Based Cloud technique, to an average reduction of 43.3% when compared to All Clouds technique. Furthermore, the percentage reduction in the number of examined services achieved by Cloud-SEnergy ranged from 50% when compared to Smart Cloud and average of 82.4% when compared to Base Cloud. In term of run-time, Cloud-SEnergy resulted in average reduction which ranged from 8.5% when compared to DC-Cloud, to 28.2% run-time reduction when compared to All Clouds.

Item Type: Article
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Computer Science
Publisher: Elsevier
Date Deposited: 29 May 2018 10:02
Last Modified: 08 Oct 2018 12:11
DOI or Identification number: 10.1016/j.suscom.2018.05.011
URI: http://researchonline.ljmu.ac.uk/id/eprint/8726

Actions (login required)

View Item View Item