Not only do data centres need to meet security, energy efficiency, business continuity and productivity optimisation needs, they also need to accommodate the unpredictable future loads on the facilities.
The world’s largest data centres can contain tens of thousands of IT devices. This is only expected to increase as our demand for digital devices and services expands. However, as data centre owners and operators have been cognisant of this rising demand and exploring ways to incorporate greater efficiencies, through consolidation and other means, future load requirements are uncertain.
The dilemma that many data centre owners and operators therefore face with catering for uncertain future loads is that it can lead to stranded plants (i.e. wasted infrastructure and underutilised equipment). Technology, as well as systems and services upgrades, leads to data centre owners investing capital in equipment that is not needed for several years, and this low usage equipment needs to be maintained, adding to expenses.
Besides the high costs for the equipment that was purchased and maintained, stranded plants are also often characterised by equipment that runs inefficiently, which leads to wasted electricity, rising power costs and it also impacts a data centre’s sustainability targets.
Currently, existing legacy data centres go through the lifecycle of upgrading systems such as Uninterrupted Power Supply (UPS), generators, incoming supplies and generators before they are faced with hard limitations such as space constraints, structural limitations and site power constraints, which lead them to quickly run out of space. In order to expand, many data centre owners make the business decision of building a new facility to overcome these limitations.
However, a staged growth approach can be more economical to manage a data centre’s current needs as well as plan and expand according to future needs.
Staged growth, as opposed to a full install, offers numerous benefits. There’s greater visibility of capital expenditure needs and lower capital costs, as well as a lower total cost since less is spent on maintenance.
Even when planning for future load requirements, however, most data centre owners and operators can still only make educated guesses regarding what the real operational growth will be.
Even a staged approach could lead to stranded assets. The problem with underutilised assets, or assets that are not used at all, is that it is usually only detected when the data centre is already in operation, making it a difficult problem to rectify. Planning the staging can help avoid stranded assets at the design phase instead of when the data centre is operating, which can lead to significant savings.
Technology selection, site location, the appointment of specialist architects and engineers, as well as planning applications, are only a few of the aspects that need in-depth consideration to build data centres for future processes. To avoid stranded assets, data centre owners also need to plan for flexibility.
When developing detailed upgrade strategies to enable modular plant expansions and provide plant expansion spaces to be added, it’s important not to skimp on building space during the design phase. Another way to plan for flexibility is to include different density zones in the site plan and provide flexibility in white space zones. It is also vital to consider all systems and services during the planning and ensure expansion capability is adequately designed with interfaces and connections.
With the emergence of digital engineering and the widespread use of Building Information Modelling (BIM), not only will modular design and prefabricated construction become more prevalent, but new builds and the operation of these will become more efficient.
Phil Motteram , leads the design of large and complex building projects for Aurecon clients across Victoria and South Australia, specialising in data centres and essential social infrastructure. Phil previously led Aurecon’s data centre business in Asia, leveraging over 20 years’ experience in the design of highly reliable, energy efficient data centres.
Please change your browser to one of the options below to improve your experience.