Sustainable data center in smart cities: the role of sustainability-related metrics

Autori
M. Chinnici, A. Grishina, A.L. Kor, E. Rondeau
Data pubblicazione
17/09/2018
Fonte
Proceedings of the 1stSustainable Solutions for Growth conference - BOOK OF ABSTRACTS, September 17-18, 2018, Wrocław, Poland
35-36

In the context of smart cities, ‘smartness’ is achieved via Information and Communication Technologies (ICTs). They offer new interdisciplinary opportunities to improve services, integrate different infrastructures while reducing resource consumption, greenhouse gases (GHG) and other polluting emissions. Meanwhile, ICT itself consumes power and has an environmental footprint and in this scenario, being energy-hungry by their nature, DCs form (energy-hungry for their nature) the integral part of global ICT infrastructure. Therefore, the role of DCs within the society is leading to increasing interest in both ICT and energy sectors. The concept of energy efficiency is becoming a complex issue in DCs, both at the design and operational stages, due to high-energy prices and policy pressures. Moreover, the influx of large data sets (Big Data) and their intensive processing demands have led to an increase in energy consumption, and undeniably, electricity usage contributes to the highest proportion of costs in DCs. A sustainable DC should entail exploitation of renewable energy, and optimisation of energy consumption without impacting the level of QoS provided to users. Hence, a sustainable DC offers an excellent opportunity to meet both environmental and business goals. The most important driver of a sustainable DC is embodied within its energy efficiency strategy. The baseline for the development of this strategy is a structured measurement/metric framework that can be used to quantify DC energy efficiency and provide insights into ways for improving it. Even though a comprehensive list of energy metrics has been presented in existing literature, limited progress has been made concerning DCs energy efficiency measurement recently. Therefore, the analysis of energy efficiency in DCs, through a set of globally accepted metrics, is an ongoing challenge. The purpose of this work is to present a comprehensive yet critical analysis of energy metric, provide a set of current DC sustainability-related and productivity metrics, followed by formalizing the role of metrics in the achievement of sustainability and its associated enhancement in the context of urban living standards. The present work discusses and analyzes the role and importance of rigorous, formal, and efficient energy management in DCs and its associated metrics to achieve this goal. This work is mainly motivated by two driving factors. Firstly, it is the prioritised focus on reducing DCs energy costs and improvement of their energy efficiency through the use of a structured measurement/metric framework and policies. The second factor relates to the following issues: availability, resource allocation (linked to SLAs), and energy consumption of smart applications that run daily in DCs. An optimised trade-off between performance and energy consumption poses a future DC sustainabilityrelated challenge. In particular, the following novel contributions will be addressed: future development of energy efficiency metrics for DCs; application of productivity metrics to resource management in urban environment paradigms (such as edge computing); suitable emerging sustainability metrics in the context of smart cities. DCs that prioritise reduced energy consumption and carbon emissions have more control over the market growth. With more sustainable DCs, IT organisations can better manage increased computing, network, and storage demands while lowering their energy costs followed by reducing the total cost of ownership (TCO) of its IT equipment. Undoubtedly, active promotion and dialogue with industrial (and Standardization) organisations concerning these metrics frameworks and methodologies, will ensure this proposed research.