Three major consumption sources of data center PUE value

The data center PUE value is an indicator of the energy efficiency of a data center, which is the ratio of all the energy consumed by the data center to the energy used by the IT load.

PUE= Total power consumption of data center equipment/power consumption of IT equipment, PUE is a ratio, the benchmark is 2, the closer to 1 indicates the energy efficiency level is better.

According to authoritative data, in the composition of data center operating costs, electricity costs account for 70%, and reducing energy consumption is to reduce power consumption to a large extent.

The lower the PUE value, the lower the electricity bill.

For example, taking a data center of 1000 cabinets as a model, if the power supply of each cabinet is 20A, the PUE value is reduced from 1.8 to 1.6, and the electricity price is calculated at 1 yuan /kwh, the electricity cost can be saved by 7.7 million yuan per year.

So, what are the top three sources of consumption that affect the PUE value in the data center?

Analysis from TGG shows that the top three areas of data center power consumption are room air conditioning, IT equipment and UPS power supplies, which together account for 93% of the total power consumption of data centers.

In order to reduce energy consumption, many places in China have introduced data center PUE value access thresholds, and data centers that exceed the local PUE threshold will be rectified and optimized to meet local standards.

If the data center is located in the suburbs, in addition to using the natural environment to cool down, the measures that can reduce power consumption are mainly to reduce the power consumption of IT equipment.

For the power consumption of IT equipment, in addition to reducing the power consumption of the equipment itself through performance optimization, an important means is to digitize the asset management method through the IT asset management system.

For example, the U-level asset management system can reduce the response speed of faulty equipment, improve operation and maintenance efficiency, accurately predict cabinet temperature changes, and then reduce the power consumption of equipment through manual intervention. For example, in a cloud data center with a cabinet as the management unit, power consumption can be reduced by shutting down faulty devices.

In a large data center, in order to reduce the difficulty of operation and maintenance, the cabinet is often the management unit, and the entire cabinet will be replaced only when a large area of cabinet equipment fails. When a few devices in the cabinet fail, the operation team chooses not to deal with it and continues to let the faulty devices run on power.