A review of energy efficiency evaluation metrics for data centers

Xiaotong Shao, Zhongbin Zhang*, Ping Song, Yanzhen Feng, Xiaolin Wang

*Corresponding author for this work

    Research output: Contribution to journalReview articlepeer-review

    56 Citations (Scopus)

    Abstract

    The rapid development of the Internet requires data center infrastructures to be extended to a large scale with a higher power capacity and a reduction in energy consumption and carbon emissions. It is of great significance for this application to reduce energy consumption, capital cost and carbon emissions while ensuring a safe and reliable data center operation. Evaluation metrics have been considered an effective method to assess the energy efficiency of data centers and have been widely used in various data centers for many years. This work reviews the latest progress of data center evaluation metrics development in recent 20 years. Energy efficiency metrics are summarized based on energy conservation, eco-design and data center security, with the advantages and disadvantages and their correlations discussed. The applicability of each metric is analyzed. The influencing factors of energy efficiency of data centers discussed include site selection, IT equipment, power supply and distribution systems, cooling systems, building structure, and temperature control. The results show that the current evaluation metrics are valuable for the assessment of energy saving of data centers. However, more comprehensive and multi-scale evaluation metrics combining energy consumption, eco-design and data center security are needed to accelerate the progress towards carbon neutrality. In this paper, the hierarchy analysis of metrics is carried out according to sustainability, and metrics in line with sustainable development should be used preferentially.

    Original languageEnglish
    Article number112308
    JournalEnergy and Buildings
    Volume271
    DOIs
    Publication statusPublished - 15 Sept 2022

    Fingerprint

    Dive into the research topics of 'A review of energy efficiency evaluation metrics for data centers'. Together they form a unique fingerprint.

    Cite this