Key Elements of a Data Center Growth Model

This audio was created using Microsoft Azure Speech Services

Ask a business leader where his company will be in the future and he’ll be able to articulate expectations about plans for growth. But predicting a company’s future with absolute certainty is next to impossible. That uncertainty, coupled with ever-changing IT and business requirements, adds to the challenge of planning a data center.

Capturing growth information in a way that facilitates the development of an IT system to serve the company’s future needs is essential when undertaking a data center project. By adopting simple planning strategies and making the uncertainty part of the equation, a company can avoid unnecessary costs while developing a suitable capacity plan for a data center or network room.

One helpful tool in formulating a plan is the growth model. The growth model provides a framework and language to expedite discussions about future IT requirements and associated power needs. The growth model includes the IT load profile (the power required by the IT load) as well as the system capacity plan (the power capacity to be provided by the physical infrastructure). The IT load profile – or growth plan – is a crucial planning element and one of three parameters used to design the power and cooling system.

There are six parameters in the growth model. The first four – maximum final load, minimum final load, initial load and ramp-up time – comprise the IT load profile. The last two – step size and margin – relate to the system capacity plan.

The IT load profile contains input provided by the user in the planning process. The system capacity plan is provided by the planning process as output to the user. It takes into account the amount of power that can be supplied to the load as well as the ability of the cooling system to cool the load

1.     Maximum final load

The maximum final load refers to the highest load that can be reasonable expected for an organization, taking into account the end user’s business plan and future opportunities. In most cases the maximum load is never reached, but some users may want to “pad” the estimate with regard to non-scalable elements such as room size to avoid unchangeable limits on future expansion.

2.     Minimum final load

This is the lowest final load that can be reasonably expected, taking into account business risk and potential changes to the market. This variable is used later to determine the initial capacity of the system as well as the phase-in plan.

3.     Initial load

This figure represents the IT load that must be supported at the initial installation. The figure relates to present business conditions and is typically less than either the maximum or minimum final load.

4.     Ramp-up time

The ramp-up time is the expected amount of time between when the system is first installed and the final load.

5.     Step size

Step size is a feature of the system capacity plan, as opposed to the IT load profile. It takes into account the scalability of the reference design (the selected system architecture), the floor plan for the data center room and an assessment of growth uncertainty and the potential total cost of ownership benefits of incremental deployment.

6.     Margin

The margin is the “extra” infrastructure capacity built in to cover unexpected changes. An unexpected change could be an increase in the power demand of the IT load (such as the unexpected addition of servers) or an unanticipated drain on system capacity (such as a decrease in cooling caused by a clogged heat rejection pipe).

Once the IT profile has been developed, the system capacity plan is formulated to support the IT load profile. By using incremental phase-in steps, the user has the option to delay, adjust or cancel full build-out based on actual conditions as they develop during ramp-up time.

For more information on the growth model and the benefits of stepped phase-in, see the Schneider Electric White Paper, Data Center Projects: Growth Model.

Tags: , ,