WUE: measuring water use in data centers
While PUE focuses on electrical energy consumption, WUE looks at water use in the data center, an equally crucial aspect, especially for cooling IT equipment. It is calculated by dividing the amount of water used for cooling by the amount of energy used by the servers. WUE ratios are expressed in liters of water per kilowatt-hour (kWh) of electrical energy consumed. The lower the WUE, the more efficient the data center is in using water to maintain an adequate temperature inside its premises.
While PUE focuses on electrical energy consumption, WUE looks at water use in the data center, an equally crucial aspect, especially for cooling IT equipment. It is calculated by dividing the amount of water used for cooling by the amount of energy used by the servers. WUE ratios are expressed in liters of water per kilowatt-hour (kWh) of electrical energy consumed. The lower the WUE, the more efficient the data center is in using water to maintain an adequate temperature inside its premises.
WUE |
Level of Efficieny |
3,0 L/kWh |
Very inefficient |
2,0 L/kWh |
Inefficient |
1,0 L/kWh |
Average |
0,5 L/kWh |
Efficient |
0,2 L/kWh |
Very efficient |
0,0 L/kWh |
Ideal |
Why is it important to measure WUE?
Many data centers rely on cooling systems to prevent servers and equipment from overheating, which can require significant quantities of water. Poor management of this resource can have a major environmental impact, especially when water becomes a scarce and precious resource.
Many data centers rely on cooling systems to prevent servers and equipment from overheating, which can require significant quantities of water. Poor management of this resource can have a major environmental impact, especially when water becomes a scarce and precious resource.
There are several ways to reduce WUE:
There are several ways to reduce WUE:
-
1. Efficient cooling technologies: The use of
more efficient cooling systems, such as
evaporative or free cooling systems, reduces dependence on water for cooling.
-
2. Waste heat recovery: Heat generated by servers can be recovered and used for useful purposes, such as heating nearby buildings. This reduces the need for further cooling of ambient air.
-
3. Reuse of cooling water: Water used for cooling can be recycled and reused, reducing the overall demand for water in the data center.
Why is it important to monitor the WUE of a data center?
A crucial aspect of reducing WUE is continuous monitoring of cooling systems and water use. This enables inefficiencies to be identified and corrective action taken to optimize water use.
A crucial aspect of reducing WUE is continuous monitoring of cooling systems and water use. This enables inefficiencies to be identified and corrective action taken to optimize water use.
According to a report by the US Department of Energy, the WUE of the average data center in the USA is 1.8 liters per 1 kWh. This means that for every kilowatt-hour of electrical energy consumed by IT equipment in a data center, 1.8 liters of water are used for cooling. Data centers with low water consumption would have a WUE of 0.2 L/kWh, equivalent to one glass of water for every kilowatt-hour delivered to the servers.
In recent years, according to Alliancy :
According to a report by the US Department of Energy, the WUE of the average data center in the USA is 1.8 liters per 1 kWh. This means that for every kilowatt-hour of electrical energy consumed by IT equipment in a data center, 1.8 liters of water are used for cooling. Data centers with low water consumption would have a WUE of 0.2 L/kWh, equivalent to one glass of water for every kilowatt-hour delivered to the servers.
In recent years, according to Alliancy :
-
Scaleway reports a WUE of 0 for three of its data centers and a WUE of 0.15 L/kWh for another data center;
-
OVHcloud displays WUE indices of between 0.17 and 0.20 L/kWh depending on its data centers;
-
Meta announces a WUE of 0.26 L/kWh;
-
Microsoft reported a WUE of 0.49 L/kWh globally and a WUE of 0.1 L/kWh for the EMEA zone (Europe, Middle East and Africa).
Measuring data center energy efficiency
It's important to note that WUE, like PUE, should not be looked at in isolation. In combination with other indicators, it offers a holistic view of a data center's energy and environmental efficiency.
It's important to note that WUE, like PUE, should not be looked at in isolation. In combination with other indicators, it offers a holistic view of a data center's energy and environmental efficiency.
Choosing an eco-responsible data center
Choosing a data center located in a low-carbon footprint region, such as France, can also help reduce the overall environmental impact of Cloud services. By supporting data centers powered by renewable energy sources and adopting sustainable practices, businesses and users can play a key role in reducing the carbon footprint of the data center industry.
Choosing a data center located in a low-carbon footprint region, such as France, can also help reduce the overall environmental impact of Cloud services. By supporting data centers powered by renewable energy sources and adopting sustainable practices, businesses and users can play a key role in reducing the carbon footprint of the data center industry.
Aqua Ray aims for carbon neutrality. In our article, we announced an efficient PUE of < 1.39 for our data center in 2022. In terms of WUE, we're on a high level of efficiency since we don't use water to cool our computer rooms.
Aqua Ray aims for carbon neutrality. In our article, we announced an efficient PUE of < 1.39 for our data center in 2022. In terms of WUE, we're on a high level of efficiency since we don't use water to cool our computer rooms.