Predicting the Future for Data Center Growth Made Easy

Predicting the Future for Data Center Growth Made Easy

This piece was originally published in the July 2016 issue of electroindustry.

Patrick Donovan, Senior Research Analyst for the Data Center Science Center, Schneider Electric

data center (2)

Successful navigation through the planning challenges of a data center project requires the structure of a well-defined process, the expertise of people to make decisions and evaluate alternatives, and the assistance of tools to organize information or perform calculations.

One such growth model that provides structure and terminology for the discussion of future IT power requirements is detailed in a white paper by Schneider Electric, Data Center Projects: Growth Model. Having a shared understanding and reasonably accurate prediction of how IT and business needs will grow and change over time is critical. Poor planning may result in wasted investments or a dramatically reduced site lifespan.

Predicting the future does not mean using mystics, crystal balls, or wild guesses. A good growth model provides project teams with a structured framework that ensures that the right information is considered in the right order by the right people to help make this less art and more science. A growth plan is one of three IT parameters in planning a data center project; the others are the criticality of the IT load and the maximum buildout capacity for the load.

The further into the future an IT load projection is extended, the less confidence there is in it. Thus, adding a value for an initial starting load, followed by the ramp-up time to the minimum and maximum final loads, completes the growth projection, as illustrated in figure 1 below.

data center chart (1)
Figure 1. Maximum load, minimum load, initial load, and ramp-up time make up the first four parameters that comprise the overall growth model.

The final element of the growth model is the system capacity plan, which is the planned deployment of the power and cooling infrastructure to support the projected IT power load.

The system capacity plan is determined after the details of the system architecture and the physical space become known. Figure 2 shows the complete growth model, composed of six parameters in total; the first four are explained in figure 1. The final two parameters make up the system capacity plan. Step size (the fifth parameter) refers to the incremental step (or block) size of the physical infrastructure systems (typically expressed in kW or MW of capacity), assuming that a full buildout is deferred.

Infrastructure systems today are much more optimized for modularity, making it much easier to do buildouts in a pay-as-you-go process, saving on both capital expenditure and operational expenditure. The final parameter is the margin, which refers to the extra capacity to cover unexpected addition to the IT load or an unexpected peak draw on bulk capacity.

data center chart (2)
Figure 2. Complete simplified growth model.

An author of numerous white papers, industry articles, and technology assessments, Mr. Donovan has conducted research on data center operations and physical infrastructure technologies that offers guidance and advice on best practices for planning, designing, and operation of data center facilities.

Read this month’s issue of electroindustry.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.