Comprehensive Guide: How to Measure Electrical Energy in a Data Center

In the dynamic world of modern technology, data centers play a crucial role in powering the digital infrastructure that drives our daily lives. Measuring the electrical energy consumption within these data hubs is essential for optimizing efficiency, reducing costs, and minimizing environmental impact. This comprehensive guide will delve into the intricacies of measuring electrical energy in a data center, equipping you with the knowledge and tools to become a true expert in this field.

Understanding Power Usage Effectiveness (PUE)

The primary metric used to measure the electrical energy efficiency of a data center is the Power Usage Effectiveness (PUE) metric. PUE is a ratio that compares the total energy consumed by a data center to the energy used solely by the IT equipment. The formula for PUE is:

PUE = Total Facility Energy / IT Equipment Energy

To calculate the PUE, you need to accurately measure the total energy consumed by the data center, including lighting, cooling, and other non-computing functions, as well as the energy used specifically by the IT equipment.

For example, if the total facility energy is 200 kW and the IT equipment energy is 94 kW, the PUE would be:

PUE = 200 kW / 94 kW = 2.13

A lower PUE indicates a more energy-efficient data center. The industry benchmark for average PUE of data centers worldwide is around 1.58, but leading organizations are striving for even lower PUE values.

Measuring IT Equipment Energy

To measure the energy consumption of the IT equipment, you can use power distribution units (PDUs) that provide detailed output measurements. Newer PDUs can offer readouts of each electrical phase, allowing for ongoing monitoring and tracking of efficiency gains.

When measuring the IT equipment energy, it’s important to consider the following factors:

1. Power Consumption: Measure the power consumption of individual servers, storage systems, and network devices in watts (W) or kilowatts (kW).
2. Energy Consumption: Calculate the energy consumption by multiplying the power consumption (in watts) by the operating time (in seconds). For example, if a server consumes 100 watts and operates for 1 hour (3600 seconds), the energy consumption would be 360,000 joules.
3. Power Factor: The power factor is the ratio of real power to apparent power, and it can impact the accuracy of energy measurements. Ensure that your PDUs can measure and account for the power factor.
4. Electrical Phases: In a three-phase electrical system, measure the energy consumption of each phase separately to identify any imbalances or inefficiencies.

Measuring Total Facility Energy

To measure the total energy consumed by the data center, including the non-IT equipment, you can use various metering techniques:

1. Transformer Metering: Install clamp-on meters on the uphill side of the transformer to measure the total facility energy, including detailed readouts of each electrical phase.
2. Automatic/Static Transfer Switch (ATS/STS) Metering: The output of the ATS/STS provides an optimal metering point for the facility’s power, especially in environments with a backup generator.
3. Building Management Software: Utilize building management software that continuously monitors and reports the total power consumption of the data center, making it easy to access and track the facility’s energy usage.

Data Center Infrastructure Efficiency (DCiE)

In addition to PUE, you can also use the Data Center Infrastructure Efficiency (DCiE) metric, which is the reciprocal of PUE:

DCiE = IT Equipment Energy / Total Facility Energy

A higher DCiE indicates a more energy-efficient data center. The DCiE metric provides an alternative way to measure the data center’s energy efficiency, focusing on the proportion of energy used by the IT equipment compared to the total facility energy.

Energy Intensity Metric

Another useful metric to consider is the energy intensity, which measures the amount of energy consumed per unit of work output. This metric can provide valuable insights into the organization’s compute needs and the efficiency of converting compute capabilities into tangible value.

The energy intensity formula is:

Energy Intensity = Total Facility Energy / Work Output

The work output can be measured in various ways, such as the number of transactions processed, the amount of data stored or transferred, or the number of virtual machines (VMs) or containers deployed.

By tracking the energy intensity over time, you can gain a deeper understanding of the data center’s energy efficiency and identify opportunities for optimization.

Practical Considerations

When implementing a comprehensive energy measurement system in a data center, consider the following practical aspects:

1. Sensor Placement: Strategically place sensors and meters to capture accurate energy consumption data, ensuring they are installed at the appropriate points in the power distribution system.
2. Data Collection and Analysis: Establish a robust data collection and analysis framework to gather, store, and interpret the energy consumption data. This may involve integrating with building management systems or developing custom monitoring solutions.
3. Benchmarking and Trend Analysis: Compare your data center’s energy metrics (PUE, DCiE, energy intensity) against industry benchmarks and track trends over time to identify areas for improvement.
4. Optimization Strategies: Leverage the energy consumption data to implement targeted optimization strategies, such as load balancing, server virtualization, or infrastructure upgrades, to enhance the data center’s overall energy efficiency.

Conclusion

Measuring electrical energy in a data center is a crucial step in optimizing efficiency, reducing costs, and minimizing environmental impact. By understanding the PUE, DCiE, and energy intensity metrics, and utilizing various metering techniques, you can gain a comprehensive understanding of your data center’s energy consumption and take informed actions to improve its performance.

This guide has provided you with the necessary knowledge and tools to become a true expert in measuring electrical energy in a data center. Remember to continuously monitor, analyze, and optimize your data center’s energy efficiency to stay ahead of the curve in the ever-evolving world of digital infrastructure.

Reference:

1. Best Practices Guide for Energy-Efficient Data Center Design: https://www.energy.gov/sites/default/files/2014/03/f6/f4605-010_0.pdf
2. Measurement and Monitoring in Data Centers: https://www.42u.com/measurement/pue-dcie.htm
3. Understanding Data Center Energy Intensity: https://www.thegreengrid.org/file/432/download?token=rDHFXQUs