Where Will Your Data Center Be in 20 Years?

Where Will Your Data Center Be in 20 Years?

A data center that is designed with flexibility really doesn’t have a life expectancy.

John M. Hawkins is VP of Corporate Marketing and Communications for vXchnge.

Twenty years in technology may seem more like 100 years when compared to other industries. In just one year a company’s landscape can change significantly. Think about how businesses scale and operate on a functional level, then add in changing technologies along with the exponential increase of data and dynamic content needed to drive business.

Will your data center strategy survive 10, or even 20 years? Will the company grow like your key stakeholders expect? If so, you may need multiple data centers, strategically located, just to handle your requirements. On the other hand, your CFO might have a more conservative estimate and is responsible for how much is actually spent on a data center(s).

In addition to size, you have to consider whether your data center(s) might become obsolete in 5, 10, 15, or 20 years.

According to Pitt Turner, executive director with the Uptime Institute, there is no set lifespan to a data center. “A data center that is designed with flexibility really doesn’t have a life expectancy,” says Turner. “Over the life of the data center, you need to replace the capacity components just like you replace the tires on your car.” Turner further states, “Chillers, UPSes, that sort of stuff, needs to be replaced and you have to have a data center infrastructure that will allow you to do that.”

Software-Defined-Data-Center Platforms

The Internet-of-Things and cloud will no doubt have a big impact on the data center in the future. We will see this throughout the various layers in the data center stack from the carriers, up through the facilities, and continuing to the data or business layers.

In the future we will see a tighter cooperation and coupling between the logical layers in a data center. So rather than each of the layers (i.e., carrier, facilities, hardware, software) working in silos, we will see more integration and ultimately the need for more orchestration and automation.

Network Optimization

Data centers are the physical location where the cloud (IaaS, PaaS and SaaS) calls home. There is a current technology gap, which will need to be solved to help support the needs of tomorrow’s businesses.

The current challenge is that a network is like a chain, where the weakest link can cause performance issues even on the most finely tuned networks. The networks of tomorrow will need to leverage multiple carriers to provide network redundancy.

These backbones will most likely be stretched layer 2 networks that are optimized for speed and security that will provide low latency high-speed networks for applications. Looking down the road, these technologies will become mainstream.

Automation is the Key

Many of the automated tasks that we see in today’s data centers will be completed through automation. This could be assumed in the Software-Defined-Data-Center paradigm, however, the type of automation that we will see in the data centers has more to do with the automation of business-related tasks and activities. Process engines will continue to get smarter and smarter to the extent that they will be able to handle complete decision trees – almost like an artificial intelligence of sorts for the data center.

The automation process engines combined with the newer optimized networks will give integrated redundancy and failover capability unseen today. This will provide more and more flexibility as companies build their N+1 data center "edge" strategy.

Consider Location

Application developers are leveraging federated design models to build applications which are having an impact on data center design and selection. For example, instead of storing massive amounts of data in a few select data centers, application providers are moving their applications to “the edge,” (in locations where they can serve customers locally, and reach more businesses and consumers in more markets) in order to be closer to the consumer to reduce latencies and perform at higher data transfer rates.

Having a data center partner that has reliable and resilient infrastructures strategically situated in multiple areas around the country will help keep data safe and processing properly.

Planning For the Future: Data Center as a Service

Another way of looking at the future of your data center is to consider your data center as a service. Here, you don’t have to allocate capital on buying real estate. You simply use a colocation provider that has already made the investment – and a plan to maintain the data center now and many years from now, offering flexibility to support future growth.

Data center as a service also offers plenty of added value via in-house services as well as access to partner services that your company can easily plug into. In addition, data centers that have a diverse and abundant environment of carriers will offer more options than facilities that lack such an ecosystem. Today, more cloud, hybrid and even on-premise applications are relying on a dual carrier network strategy.

Conclusion

You really have two options when considering your long-term data center strategy. First, you can use your own research to guestimate the needs of your company, taking on a huge capital expense while building your own data center.

The second option is to consider a colocation service provider that has already done the research, is building and innovating, offers a variety of locations, services and products to leverage, and is optimized and ready for growth. This makes the provider responsible for upgrading expensive equipment as needed and making sure the data center is future proof.

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. View previously published Industry Perspectives in our Knowledge Library.

Hide comments

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.
Publish