Chris Wellfair, projects director at Secure I. T. Environments Ltd, looks at why a containerised data centre may be a smart move
You’ve probably heard the term ‘containerised data centre’ and associated it with huge data centre projects, such as those run by the public cloud vendors with tens of thousands of servers and the need for constant growth. In those instances, they are chosen for speed, cost effectiveness and ease of installation, but those benefits are not only true when working at scale. There are several scenarios where a containerised data centre might be just right for you, even if you are a much smaller business. Modular data centres have, for some time, helped customers lower the cost of their new facility and overcome the challenges associated with such builds.
Modular data centres can also provide secure facilities within existing buildings, or allow a data centre to be put in unusual locations where it would not be appropriate to have a new build, either because of planning laws or space.
Containerised solutions can form small data centres where they help resolve space, deployment time, build complexity and cost challenges. But they are very flexible and depending on the internal configuration can perform very well in high density applications.
In an emergency, containerised solutions can also shine as part of a disaster recovery plan if already fitted out with the infrastructure, where they can be rapidly deployed to a site as a temporary solution.
If you are facing a data centre design and build challenge and any of the following resonate with you then you should certainly add a containerised data centre to the possible solution list.
A container can be a solution accepted by all and can be implemented with a minimum of fuss or raised eyebrows from the CFO
Speed is of the essence so if you are looking for your data centre to be built quickly, containerisation can substantially shorten delivery times. Many companies offer them in standard ‘ready to load’ configurations, but you can, of course, have the interior designed to meet specific requirements, if your partner offers this.
Site suitability also needs to be a consideration. In some locations, it is simply impossible to house a new data centre. This could be due to footprint, budget or even local planning regulations. Often in these situations, a container can be a solution accepted by all and can be implemented with a minimum of fuss or raised eyebrows from the CFO.
There could also be many reasons why you can’t build a data centre on site, for example, if it is a high security area, or the data centre is only needed in a disaster recovery situation such as a flood, so you want to keep it offsite. A containerised solution can be fully designed, fitted out and tested at a separate location. It could even be running in a separate location mirroring servers at the main location, and can then dropped in as a ‘clone’ when needed.
Additionally, if you need your data centre to be mobile, either because you know the facility will be moving to another site in the future or you want to ship it to another country once built, then containerisation is an excellent solution. Firstly, because units are designed to shipping container external dimensions, use the same interlock systems and meet or exceed the same rigidity and load standards, shipping them is a lot easier than sending individual components that must be re-assembled at the other end. Secondly, it is possible to get the container insured if it meets the correct international shipping container standards, giving you greater peace of mind.
There are a couple of misconceptions about containerised data centres, the first being that they really are just a temporary solution and have a touch of the Heath Robinson’s about them. To a degree, this is understandable, after all they do look like an upcycled shipping container, but the technology in them is the same as would go into ‘normal’ data centre build, or at least from the same suppliers. If you pick the right partner, then your container will be custom designed and built from the frame up and will carry enviable Lloyds Register structural warranties to give you peace of mind. You’ll have extended or upgraded the data centre long before those warranties expire.
The other reasonable question that any data centre manager worth their salt will ask, is about the ability of a containerised data centre to maintain effective cooling and achieve strong power usage efficiency (PUE) ratings – the misconception is that they will fall short. They can actually deliver the same high standards as a modular or traditional data centre builds. As outlined above, this is because they use the same equipment, including monitoring systems – they are well suited to high density applications where heat can be an issue precisely because of the way containers are configured. Also, where there are particularly stringent demands it is not uncommon to have a second container that is responsible for housing switchgear, batteries, UPS and cooling hardware, though can be housed within container ‘rooms’.
Containerised data centres are not a replacement for a modular room or bespoke data centre build, they are simply another option. In certain situations their advantages may make them perfectly suited to the challenges that you are trying to overcome. The important thing is to consider each option on its merits and select the solution that meets both your strategic IT goals, as well as the future plans of your organisation.
This post originated at Data Centre Management magazine, from the same publisher as The Stack. Click here to find out more about the UK’s most important industry publication for the data centre space.
