Skip navigation

New Technologies Keep Data Centers Low-Rise

Once you reach outside of the most populous areas of the country, there’s much greater possibility of placing multiple centers instead of monolithic structures.

Steven Cooke is Managing Director for the USA West Region at Linesight.

As the biggest players in computing build out capacity to keep up with the exponential growth of the internet economy, it’s worth considering what the data center of the future will look like. For many years the thinking was that massive server farms would be housed in science fiction-like skyscrapers. That doesn’t seem likely, but not for the reasons one might think. Today’s computers are almost too advanced to justify the creation of massive data fortresses. That is causing technology companies to radically reimagine how to build data centers that are energy efficient, easy to maintain, and able to scale to meet user demand.

Data center capacity is growing at an incredible rate as companies like Google, Amazon and Microsoft pour investment into tools that will make their systems more efficient. Recode reports that the data giants spent $20 billion last year buying data centers, and it appears as if 2018 might be even bigger. To meet user demand, the trend seems to be the deployment of large numbers of data centers throughout the country as opposed to a few enormous structures in places like Silicon Valley and New York City.

The critical advantage of distributing data centers around the country is the availability of land at affordable prices. Once you reach outside of the most populated areas in the country, there’s a much greater possibility of placing multiple centers instead of monolithic structures. The local data centers can also be much smaller, more like a shipping container than a 2-story building. With technology allowing computing units that small to do massive amounts of work, the incentive for spreading these smaller centers around becomes very clear.

Location, Location, Location

Placing data centers in strategic locations is important, because it allows for less lag over a much greater area. One high-rise data center, even if hypothetically possible, would in most cases not be close enough to users for the growth rate of the data economy. The only places where this is really being explored is high-density population server destinations like Singapore. But for most companies in most regions, and certainly in the United States, there’s a lot more ground to cover. Having multiple high-density data center environments is a much better approach, especially given the rise of virtualized systems. The trend of gigantic cloud companies buying everyone else’s data centers feeds into this placement-over-size approach.

Virtualization is a key part of the new data economy, and it’s a driving force in new data center design. Physical servers now supporting millions of digital machines. As this trend continues and becomes even more complex, the capacity for these systems to become more autonomous and powerful only grows. All signs point to this becoming an even more important part of the data center world, and another factor in them not growing in physical size over time. Finding ways to monitor and grow these virtual environments is much more important - and that can be better done with smaller, specialized systems instead of bigger ones.

Data centers are becoming a towering part of the technology landscape, but that doesn’t mean their footprint is growing. Companies and managers are focusing on making systems more efficient instead of just bigger. Smaller data centers are doing important work around the country, and cheaper land costs outside of big cities incentivize heavier distribution instead of more big structures. Data centers are big already, and despite the research being done by nations that have more concentrated or massive populations, it seems unlikely that high-rise data centers will catch on as a concept. Up into the cloud, but not up into the clouds.

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.


Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.