The Stack Archive Feature

Why data centre infrastructure standardisation benefits everyone

Thu 11 Jan 2018 | Patrick Donovan

Patrick Donovan, senior research analyst at the Data Center Science Center at Schneider Electric, explains the benefits of standardisation, both in terms of availability and cost.

Standardisation is a key factor driving growth and user acceptance across most industries, particularly in emerging technologies where fear of becoming locked in to a minority product line can inhibit market adoption.

Although the data centre industry is maturing rapidly, it is still in a state of high growth and the key drivers of success are speed to deployment, scalability and reliability: all features that benefit greatly from the presence of widely accepted and trusted standards.

USPs are greatly desired in many industries, but from the point of view of data centre infrastructure, familiarity is key.

Vital infrastructure elements such as fans, wiring racks, and UPS (uninterruptible power supply) systems are more valuable by virtue of being interchangeable, interoperable and requiring the least effort possible to deploy.

One-time engineering costs, although inevitable, are greatly reduced by standardisation, which minimises the effort required to have a new infrastructure element perform to specification with existing equipment.

Reducing complexity in the data centre

Efforts to bring the benefits of standardisation to data centre infrastructure deployments include the emergence of reference designs and prefabricated data centre solutions, which greatly increase the speed of deployment and reduce any accompanying integration complexities.

Modular components with standardised structure and connections make everything easier, faster, and cheaper – from manufacture and inventory at the vendor, through design and engineering at the planning stage, to installation and operation at the customer site.

Prefabrication and modular designs are the sources of one critically important component of physical business value (agility, the ability to respond to changing or unexpected business opportunities) and a major contributor to two others (availability and total cost of ownership).

For these to work together seamlessly, an adherence to standards is essential, especially if one-time engineering costs are to be kept in check.

Reference designs can encompass infrastructural elements from modular subsystems right up to prefabricated, fully-functional data centres and are based on proven best practices for the deployment of technology for a specific purpose.

Similar to blueprints, they are conceptual plans for how the physical infrastructure systems are configured and laid out and cover facility power, facility cooling, and the IT space.

Importantly, they also include a standardised description of the system or subsystem that describes all the different attributes and performance characteristics for a given set of requirements, including electrical lines, piping diagrams, floor layouts and dimensions and equipment lists.

By providing a written plan for the specification and integration of a variety of elements into a prefabricated solution, reference designs offer a reliable and cost-effective way to deploy and scale out infrastructure in a timely manner. They reduce the number of decisions that have to be made and limit the complexity of assembling integrated systems.

Such simplification of effort depends in turn on the availability of standardised equipment that can be incorporated into reference designs so that amendments can be made over time to improve performance or introduce new functional capability into an existing system.

Data centre infrastructure collaboration

Although it is very much in the interest of individual vendors to publish reference designs featuring only their own products, the nature of the data centre industry is that installations typically include best-of-breed products from a variety of specialist vendors.

For these to work together seamlessly, an adherence to standards is essential, especially if one-time engineering costs are to be kept in check.

Hyper-converged systems, which integrate many processing, storage and networking technologies into a single highly integrated unit, might seem at first glance to be the ultimate in proprietary systems: if everything essential is integrated into a single box, it shouldn’t matter how standardised or ‘open’ that integration is.

In practice, even hyper-converged solutions have to interact with other infrastructural elements such as UPS and containment systems.

As such, vendors of complementary infrastructural elements are keen to collaborate on the development and adoption of standards to promote interoperability and compatibility to ensure that their products are available to the greatest number of users.

Standards allow vendors to compete with each other on the basis of merit rather than exclusivity. In a fast growing and always developing industry such as data centre infrastructure, high performance allied to accepted industry standards is a winning combination.

Experts featured:

Patrick Donovan

Senior Research Analyst
Schneider Electric


Data Centre DCIM feature infrastructure
Send us a correction about this article Send us a news tip