A self-assessment framework for evaluating efficiency of data centers

Abstract

Increasing global data center energy consumption, with the rapid digitalization of society, has driven the need for comprehensive assessment reports periodically. Current challenges include the complexity of collecting data and calculating Key Performance Indicators (KPI) within the corresponding reporting period. To address this, we introduce the Self-Assessment Tool (SAT) as a unified, modular, and extensible framework for the evaluation of thermal and energy performance of data centers based on IT and cooling data sourced from data monitoring systems. A historical dataset can also be imported to the SAT to assess the energy efficiency of a data center based on the data collected from external sensors. The KPI calculation module of SAT calculates a comprehensive set of standardized thermal (RCI, RHI, RTI, RI, and LI) and energy (PUE and COP) metrics based on both real-time and historical datasets, with a transparent and reproducible KPI computation workflow that enables independent implementation. As demonstrated and validated on two pilot data centers located in Denmark and Switzerland, the SAT automatically generates assessment reports including time-series visualizations, rack-level thermal maps and energy-efficiency classifications according to the KPIs. Finally, the proposed framework can be deployed either as a standalone or web-based service for the performance evaluation of data centers following the present assessment workflow.