Data Center Metering and Resource Guide
Data centers constitute a large and growing sector of energy use. By one estimate, they consume about 2 percent of U.S. electricity, and the sector is increasing energy use by about 1 percent each year (Shehabi et al 2016). Inefficiencies in the power and cooling systems of the data center infrastructure often give data center owners and operators significant opportunities for energy-efficiency measures. Taking advantage of such opportunities typically is very cost-effective, reducing the operating costs, and can also improve data center reliability. Managing and assessing energy performance requires the use of established data center energy performance metrics, and metering is needed to provide the data for such metrics.
Tracking and managing data can inform decision makers in the following ways:
- It helps organizations identify abnormally low- or high-energy usage and potential causes, supporting such practices as peak-shaving. It also facilitates capacity planning around space and power utilization and helps with carbon accounting and greenhouse gas (GHG) reporting.
- It helps organizations track and manage energy costs, verify energy bills, and prioritize, validate, and reduce energy costs through improved energy efficiency and energy management.
- It allows organizations to quantitatively assess data center performance and to benchmark it across a level playing field. Benchmarking evaluates the organization's position relative to the rest of the market (cross-sectional benchmarking) or over time in one data center (longitudinal benchmarking). This enables engagement with senior management and other stakeholders to participate in continuous improvement of the organization's energy performance.
- It helps organizations develop and validate energy-efficiency strategies and identify opportunities to improve energy efficiency by lowering energy and operational costs. These strategies include identifying large energy users and establishing performance metrics (cooling plant kilowatts/ton, airhandling watts/cubic foot per minute, etc.) for monitoring and tracking.
- Energy performance metrics can also be used to commission and detect faults in physical systems and diagnose their causes.
This guide intends to help data center owners and operators implement a metering system that allows their organizations to gather the necessary data for effective decision-making and energy-efficiency improvements.