Cooling, air conditioning and energy for computing (most likely in that order) are the key factors that determine ongoing energy costs in every datacenter. IT Hardware is not only expensive to run but also expensive to purchase (not to mention UPS, generators, the building itself, lighting etc.) and makes it very costly to build a new datacenter or refit an old datacenter or repurpose an existing property. With traditional air-cooled datacenter, maintaining low levels of humidity and temperature (not to mention the not so obvious benefits of liquid immersion cooling) is extremely important to ensure correct operation of the IT Hardware.
Power Usage Effectiveness (PUE) and its reciprocal datacenter Infrastructure Efficiency (DCiE) were introduced by the Green Grid, a nonprofit made up of a consortium of various disciplines (technology providers, facility architects, end-users, utility companies and policy makers) that collaborate to improve the efficiency of datacenters.
Using PUE as a measurement helps understand how efficient a datacenter is and compare with similar datacenters in similar locations or with similar environmental conditions, to determine whether there are areas that could be improved by adopting new technology and by applying best practises and architectural choices.
PUE represents the ratio of total amount of energy used by a computer data center facility to the energy delivered to computing equipment:
PUE = Total energy entering the datacenter / Energy used by IT equipment inside the datacenter
DCiE is a metric used to evaluate the power or energy efficiency of a datacenter. DCiE represents the ratio of total amount of energy consumed by all IT equipment and resources to the entire energy consumption of a datacenter.
DCiE = 1 / PUE
The following table gives you an example of the relationship of PUE and DCiE (the typical value of PUE lies between 1.2 and 2.5, and the DCiE is inversely proportional to the PUE value).
To be a meaningful benchmark, PUE/DCiE should be measured on a regular basis and also on different days of the week and at different times of the day. In this way, you will be able to check if adjustments you made to your datacenter improved your energy efficiency or not (here are some examples of online PUE calculators).
Considering the increasing costs of energy, most datacenter administrators are more and more focusing their attention to optimize the power usage of their datacenters measured by using PUE and DCiE.
The PUE of a datacenter can ben affected by some variables that are specific for each site:
A recent survey by Supermicro shows that there is still a general tendency to underestimate the real importance of the PUE for datacenters:
According to survey results, the overwhelming majority of businesses don’t prioritize power consumption of the equipment when they develop datacenter strategies, despite the fact that more efficient hardware would deliver energy savings over its lifetime, lowering both TCO and OPEX.
A low PUE has a clear, immediate meaning for datacenters: it indicates that the datacenter operator is not only reducing operating costs but also helping to reduce impact datacenters have on the environment by being greener, more efficient and generating less carbon emissions (as witnessed for example by the new Facebook datacenter in Singapore).
A step that our friends at the Green Grid should perhaps consider, would be to add the positive effect the re-use of heat to the PUE and DCiE equations.
Get in touch with us at firstname.lastname@example.org to find out how Submer Immersion Cooling can help you and your datacenter achieve a PUE of 1.03 or better today.