Title: An algorithm to optimise the energy distribution of data centre electrical infrastructures

Authors: Joao Ferreira; Gustavo Callou; Paulo Maciel; Dietmar Tutsch

Addresses: Informatics Center, Federal University of Pernambuco, Recife, Brazil ' Department of Computing, Federal Rural University of Pernambuco, Recife, Brazil ' Informatics Center, Federal University of Pernambuco, Recife, Brazil ' Automation/Computer Science, University of Wuppertal, Wuppertal, Germany

Abstract: For the overriding need of storing all the data that is produced by social networks, the increase in energy consumption by e-commerce and cloud computing demands critical attention. Parallel to this surge in the use of electricity there is a long held hope for a reduction in both financial and environmental costs. What is presented here, in order to optimise energy distribution infrastructures at the data centre, is the power-balancing algorithm (PLDA-D). In order to analyse energy flow models (EFMs), the PLDA-D bases itself on the Bellman and Ford-Fulkerson flow algorithms. Power efficiency, sustainability, and cost metrics of data centre infrastructures constitute the elements for EFM computation. The case study that follows analysed four power infrastructures to demonstrate practical applications for power reduction strategies. A 3.8% reduction in sustainability impact and operational costs is shown by the results obtained.

Keywords: sustainability; energy flow model; dependability; optimisation; data centre power architectures.

DOI: 10.1504/IJGUC.2020.107625

International Journal of Grid and Utility Computing, 2020 Vol.11 No.3, pp.419 - 433

Received: 19 Jul 2018
Accepted: 20 May 2019

Published online: 24 Apr 2020 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article