Evaluating the sustainability of big data centers using the analytic network process and fuzzy TOPSIS.

Abstract

The big data revolution has created data center sustainability problems, whose solutions require the consideration of environmental factors. The purpose of this study is to establish a big data center sustainability evaluation index and provide guidance for sustainable data center construction. This research formulated a big data center sustainability evaluation model that integrates multiple-criteria decision-making methods based on the analytic network process and fuzzy technique for order preference by similarity to an ideal solution (TOPSIS). Furthermore, a case study was used to examine the proposed model. The refrigeration system, layout and ventilation, data center location, data volume, and server power consumption are the five most crucial factors in determining the sustainability level of a big data center. The areas that require further development are the balancing of tasks on different IT equipment, renewable energy use, and waste heat utilization. This research provides a method or guide that can be used by managers when they build new big data centers or upgrade and optimize existing big data centers to make them more sustainable. This study is the first to assess the sustainability of a big data center according to multiple criteria decision-making methods, in which fuzzy theory is applied to evaluate the imprecise and subjective judgments of decision-makers. This study provides a systematic evaluation framework that is based on qualitative and quantitative criteria and comprises the four factors of big data level, equipment level, room level, and data center level. Big data is new oil, but it is not clean oil. It is both a vital driver of economic growth and a source of environmental damage. We need to ensure that big data centers are run in a sustainable way.