The challenge we all face today is to have consensus on how "data center efficiency" should be defined, what elements should be identified as contributors to data center efficiency and how they should be measured. This is why we have developed the OpenDCME model, which is based on their best practices. It is available with a Creative Commons license and you are welcome to use the model in your own data center to measure your current status and to start improving your data center's energy efficiency.

The model spans the data center and IT as its user. You are probably familiar with PUE, DCiE, EUE and similar methods to determine the energy efficiency of the data center. They focus on the energy consumption and it is a first trigger. But reality is more complex than that.

If PUE/DCiE/EUE or similar is your primary KPI? What happens to your KPI if

  • IT reduces the amount of hardware and therefore very likely their energy consumption drops too?
  • External temperatures are higher than average and therefore the cooling system has less benefit from its free-cooling capability?
  • IT adds new hardware, that do not fit in the contained cold/hot aisles that you installed recently?
  • You do not trend the metrics from facility and IT systems to understand how changes and other behavior impact the electrical and cooling systems?

In all cases, your KPI will drop. As said, reality is more complex. You can use the model to:

  • Reduce energy consumption and therefore reduce energy cost.
  • Zero-measure your chain of teams from the facility via the IT architects up to IT hardware procurement.
  • Identify opportunities to improve.
  • Identify relationships in the complex organization and the impact of tuning parts of it.
  • And much more.

Please proceed to the overview of the model to learn more about it.