Green data center metrics are crucial to ensure the low emission level of your data center. It is a way to measure the environmental impact of your data center and track progress towards becoming more sustainable.
You can measure carbon emissions using different tools, but you must choose an easy tool to use and understand.
This article will guide you through the process of defining green data center metrics and how to measure them.
5 Most Common Green Data Center Metrics
Data centers consume large amounts of energy to power their servers, cooling systems, and other equipment. Increasing demand for data center services has led to concerns about the impact of data centers on global warming.
To address these concerns, many companies have implemented initiatives to reduce the energy consumed by their data centers.
These initiatives include measures such as reducing the number of servers used in each facility or reducing the time that equipment remains on standby mode during periods when there is no need for it to be operational.
Here is five key metrics that can help you measure the low emission level of your data center.
1. Emission level (Green metrics)
Green Metrics is a metric that measures the number of greenhouse gases released by a data center facility over a given period.
The most common types of emissions include:
- carbon dioxide (CO2),
- nitrous oxide (NOX),
- sulfur hexafluoride (SF6),
- hydrofluorocarbons (HFCs),
- perfluorocarbons (PFCs),
- and nitrogen trifluoride (NF3).
Emission levels can be measured using reports from each piece of equipment at the facility or by using an energy management system that monitors all energy consumption from each piece of equipment and compiles it into one report for easy reference.
2. Power usage effectiveness (PUE)
PUE measures the amount of power going into a data center compared to the amount of energy used by IT equipment in a data center. The lower the PUE, the better. PUE is the most popular green data center metric, with the most optimal value for this being 1.
Power usage effectiveness is the ratio of total facility power to IT equipment power. A PUE of 1 means that 100 percent of the energy being used in your data center goes to running IT equipment.
A PUE of 2 means that 50 percent of your power is wasted on non-IT activities like cooling, lighting, and ventilation. A PUE of 3 means that 33 percent of your data center energy is wasted on non-IT activities.
PUE is defined as:
PUE = Total Facility Power / IT Load Power
A PUE of 1.5 is considered suitable for most data centers, while some newer facilities boast PUEs as low as 1.15.
3. Data center infrastructure efficiency (DCIE)
Data center infrastructure efficiency (DCIE) measures the total IT equipment power consumption divided by the total IT power consumption. It represents the fraction of electricity used for cooling and other non-IT equipment.
DCiE = IT Equipment Power/Total Facility Power x 100%
A higher DCIE means that more power is being used for cooling and less power is left over for computing, while a lower DCIE indicates that more power is being used for computing and less is being used for cooling. A good DCIE value is around 1.0 or above (1.0=100%).
The Green Grid has developed the DCIE metric, an industry organization dedicated to reducing the environmental impact of computing. The Green Grid uses DCIE as part of its LEED for Data Center Operations certification program for data centers that seek to reduce their environmental footprint through sustainable design and operation practices.
4. Water usage effectiveness (WUE)
Water usage effectiveness (WUE) measures how much water usage per unit of IT equipment performance. It’s typically expressed as gallons per kilowatt-hour (gal/kWh). A lower number means the data center uses less water than average, while a higher number means more.
This metric is calculated as follows:
WUE = (MW of Energy) / (Mgal/1000 gal of Water).
Typically, large enterprises can achieve WUEs below 1 gal/kWh, while smaller enterprises may be above 1 gal/kWh. However, many factors can impact WUE, so it’s essential to look at each organization individually.
The average water usage effectiveness for all U.S. data centers is 1.7 gallons of water per kilowatt-hour (gal/kWh). A more efficient facility might have a WUE of 1 or less; a less efficient facility might have a WUE of 2 or more.
Green data centers are designed to reduce the water used in cooling systems, which consume about half of the electricity data centers use. Technologies such as free cooling and evaporative cooling reduce water use significantly.
5. Carbon usage effectiveness (CUE)
Carbon usage effectiveness (CUE) is a metric that measures the carbon footprint of the data center. You can calculate CUE by dividing the total energy consumed by the total amount of carbon emitted.
The formula looks like this:
Total energy consumed / CO2 produced = CUE
It measures the amount of carbon dioxide emitted per kilowatt-hour (kWh) electricity consumed by the facility. Use this metric to determine how much reduction in CO2 emissions is if a facility uses renewable energy sources instead of traditional coal and natural gas-based power plants.
CUE can compare similar or different types of servers in a data center. For example, if you have two different server types running at the same load and efficiency level, one of them may have a lower CUE than another. It would indicate that it would be more cost-effective to deploy more of these types of servers rather than more of another kind.
In addition, you can use CUE in conjunction with PUE to gauge how efficient an entire data center’s operations are over time.
As we know, data centers are the backbone of cloud computing and the digital economy. Data centers are large energy consumers, accounting for 2 percent of global electricity consumption, which is equivalent to the output of 200 nuclear power plants.
In addition, the data center also emits a significant amount of carbon dioxide and other greenhouse gases into the atmosphere. Therefore, it is essential to reduce data centers’ energy consumption and emission level.