Issue link: https://cp.revolio.com/i/661564
14 | THE DATA CENTER JOURNAL www.datacenterjournal.com t hanks to a variety of eco- nomic factors, many busi- nesses brought their data back in house, but they oen built facilities to meet short-term needs without anticipating future ones. Not only is this approach costly, but it's also a drag on business. Building new data centers from the ground up takes time that could best be used by putting data to work in support of business. Additionally, from an environmental standpoint, replacing a facility within 20 years—when most buildings in other sec- tors are designed for 50 or more—is coun- terproductive to the green goals of most companies. Fortunately, there is a light at the end of the server racks. Early collabo- ration with the right design and construc- tion partners can make a difference that has implications for the bottom line. A recent flight of mine arrived (early for a change!) and pulled up to the gate. While I was at the airport, I couldn't help but wonder about the market comparison of airports to data centers. We don't build many new airports in the United States, even with the dramatic industry changes from the operations side to customer demands. Although renovations, expansions and, occasionally, reconfigurations take place, airports have been tremendously resilient. With only a handful of exceptions over the past several generations, new planes are designed, put into service and seamlessly work with existing airport infrastructure. Now consider the data center world. Our industry has existed for a fraction of the time compared with commercial aviation, but it has seen a similar amount of change in both infrastructure demands and customer expectations. We have yet to find a data center model that is as sustain- able as airports in the need for wholesale change and replacement. As it stands, businesses have gone through several cycles when it comes to their data needs. A server room may have initially seemed like a great solution until the need for more data and redundancy grew. en, in a quest to reduce overhead costs, colocated data centers gained favor. anks to a variety of economic factors, many businesses brought their data back in house, but they oen built facilities to meet short-term needs without anticipat- ing future ones. Not only is this approach costly, but it's also a drag on business. Building new data centers from the ground up takes time that could best be used by put- ting data to work in support of business. Additionally, from an environmental standpoint, replacing a facility within 20 years—when most buildings in other sectors are designed for 50 or more—is counterproductive to the green goals of most companies. Fortunately, there is a light at the end of the server racks. Early collaboration with the right design and construction partners can make a difference that has implications for the bottom line. no CrystaL BaLLs Technology has moved quickly, and everyone—from consumers to the compa- nies they work for (and patronize)—needs faster access to data than ever before. e shi is apparent in how we consume music, where the industry has moved from making larger hard drives for portable devices to relying on the cloud for storage and streaming. at example is playing out in differ- ent ways across many industries. Health- care providers need access to patient data virtually and quickly. Manufacturers need to know in real time where elements of their supply chain are while in transit. Even transportation agencies are finding new ways to use data from the cloud to ease congestion on our daily commutes. Certainly, some visionary businesses and tech leaders predicted this world, but having the infrastructure to pull it off is a challenge and will remain so. Our tech- nology-driven and technology-dependent ecosystem demands the ability to scale on demand in order to satisfy the needs of ev- ery stakeholder. Because there is no crystal ball that can accurately predict what the future looks like, we must challenge ourselves in the way we design, build and deploy the infrastructure that supports the on-demand-computing world. CoLLaBoration in Design anD ConstruCtion e traditional design-bid-build method of procuring construction delivery has done little for companies building data centers. In this setup, individual services from design to construction to procure- ment have been commoditized. On paper there are savings, but there is little value in saving half a million dollars today if you have to spend $20 million five years from now. A better model starts in the board room. CIOs would see significant benefits if they invited prospective design-build teams to learn about the longer-term goals and the needs of the company when it comes to accessible data. is is a conversation that must go beyond capacity forecasts. e design-build team should consider the data center's total cost of ownership and, ultimately, how to design a facility that can incorporate new technol- ogy over its lifespan in a sensible way. Additionally, companies must con- sider just how much infrastructure—and the energy needed to deploy it—is re- quired at a given time. Modular solutions can offer data center buyers flexibility over time and the ability to add the shiny new servers quickly and efficiently. But to achieve this goal, design cannot occur in a vacuum. Upgrading a data center need not disrupt current operations, but upgrading without interruptions takes a design that that can evolve over the life of the facility. From a pure data-storage stand- point—the entry level for any project— collaboration among designers, build- ers, equipment vendors and the owner ultimately leads to developing scalable and