The Problem With PUE
Power usage effectiveness (PUE) is the standard measurement used to determine the Energy Efficiency of a data center. PUE is determined by dividing the amount of power entering a data center by the power used to run the computer infrastructure within it.
​
The PUE ratio was adopted by the hardware manufacturers of storage devices as the best way to measure consumption. The flaw with the PUE calculation is the limited scope of what's measured.
If you want an accurate measure of power consumption you need to include all of the energy used by the data center. Using just a specific segment, (storage) fails to consider more than 2/3 of the energy that's consumed.
​
Since only a fraction of the energy is included in this metric, data centers are guilty of “greenwashing” consumption. Not including Servers and Other IT, Lighting, and HVAC from the calculation prevents you from accurately depicting true usage.
​
​
The following table explains the relationship between PUE and DCiE. Usually, the value of PUE falls between 1.2 and 2.5. The table below shows the relationship between PUE and DCiE.
We are by no means a replacement for this measurement, we are an additional component that's used to achieve efficiency. The value of the PUE is gauging wether the equipment used for storage is new versus legacy. Legacy storage equipment can use twice the power of newer storage devices.
Data centers are under pressure to lower their carbon emissions. Estimates of the greenhouse gas they emit are approximately 200 million+ metric tons of carbon dioxide in a 2019..
Factors that effect data center power optimization
1.The energy efficiency of the computing equipment in the data center – The newer equipment utilizes lower power and tackles higher workloads. The age and design of equipment also directly define its power efficiency.
​
2. The ratio of computing equipment to other equipment – If the facility houses a lot of IT equipment in comparison to other equipment it will have a lower PUE and higher DCiE.
​
3. Efficiency of cooling equipment – While the efficiency of IT equipment is important to lowering the PUE, however avoiding cooling is a mistake. It's most energy-intensive task within a data center (More than 40% of the power in a data center goes towards cooling. So efficiencies in cooling will accomplish a greater level of overall reduction than just storage or servers alone.
While far from perfect the PUE is the first standard measurement used. amigodata is not a replacement for the PUE we are focused on helping data centers achieve better efficiency through quantifiable actions.