Darren Watkins is Managing Director for VIRTUS Data Centres.
In order to cope with digital transformation and the resulting mountains of data and bandwidth requirements, more and more businesses - across all industries - are shifting servers to data centers outside of their organizations, unlocking benefits such as infrastructure flexibility, better recovery options, improved collaborative systems, workforce mobility, ease of access to public cloud operators and an overall lower total cost.
Now that the data center conundrum is no longer “build vs. buy” - complexity and capacity challenges have meant that “buy” has won that particular war - the question for many organizations has become “which partner do I trust?”, and importantly, “where are their data centers located?”.
In addition to price, scalability, uptime and reliability, location is arguably the key consideration when choosing a technology partner. A good choice of location means an optimized infrastructure and application environment, capable of reaching your entire audience, or a game changing data analytics strategy that helps you understand your operations and your customers better than competitors. On the flipside, poor location can result in unstable connections and efficiency problems.
Clearly, there’s a lot to consider when choosing a partner -- from physical security, disaster recovery, data center uptime guarantees, service levels, scalability and reliability to ongoing support and maintenance – and companies need to strike a balance between location being an important component of choice, and making it the only concern when looking for a data center provider.
Getting the Basics Right
In theory, a data center can be built anywhere with power and connectivity, but in reality, location has an impact on the quality of service that the facility is able to provide to its customers.
Connectivity, for example, is a cooperative venture which relies on proximity. Ideal connectivity depends on multiple redundant fiber connections to major bandwidth providers, and the only way to provide consistent and reliable bandwidth at the volumes required by an enterprise-grade data center is to build lots of connections to lots of different network providers.
The facilities of these providers tend to cluster together at major peering points, and when data centers are located in close geographic proximity to Internet Exchanges, or peering points, the organizations using them will benefit from low-latency and multiply-redundant bandwidth.
However, regardless of how much bandwidth a data center has access to, its customers are inescapably bound by the physics and the infrastructure of the internet i.e. data takes time to travel. Round-trip distances are usually double the geographic distance, because both the request and the response have to traverse that distance. This is important to any business; surveys consistently show that internet users are quick to drop sites with slow page load times - people want access to data instantaneously.
The round-trip time is compounded by the state of the network. Data almost never travels in a straight line between sender and recipient. Instead, it meanders through networks, routers, and switches, each of which can add latency - the closer the data center to its customers, the lower the latency.
There are a number of additional factors to consider. These include local data protection laws, tax structures, access to utilities, availability of suitable networking solutions, local infrastructure, the accessibility of a skilled labor pool, track record and existing customers or reference clients.
All these things combined make it very clear that the physical location of the data center is important. However, it’s widely discussed that, in cities at least, space is running out. There has been a boom in “mega data center” to cope with increasing demand and worldwide data center space is now at a whopping 1.94 billion square feet in . This lack of land availability and increasing costs have seen data centers moving into the suburbs.
It isn’t simply a matter of building on the edge of large cities to get best performance at the lowest cost. A knowledge of the national power infrastructure is needed to future proof any investment; an in-depth understanding of where the fiber operators’ networks exist and being able to provide an “on-ramp” access solution to public cloud platforms is critical to any enterprise deployment today.
Businesses should look for data center providers that have combined lower-cost availability of ample space and power for hyper efficient data centers with the availability of broad and rich connectivity - fiber that digital businesses need. These facilities are usually far enough from city centers for disaster recovery purposes and avoiding expensive city center premiums, but close enough to be easily accessible by local and international businesses.
Is location really everything when it comes to data center placement? In short - no. When choosing a colocation data center provider, there are many factors to consider, including connectivity, reputation, services, support and cost. But, it’s certainly true that one of the most important factors is location. Location is central not just because the geographical location matters, but because location has an impact on many of the other factors which are crucial to data center success.
Opinions expressed in the article above do not necessarily reflect the opinions of Data Center Knowledge and Informa.
Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating.