What Existing Facilities Are Best Suited to Become Data Centers?

As the world’s reliance on technology grows, so does the need for high-speed computing, data cloud storage and data centers to securely manage the world's exponentially increasing zettabytes of data. In the last two years, more data has been shared online than in previous human history.

Additional data centers are needed to physically house the large groups of computer servers and systems that support the global storage, processing and distribution of data. As a result, the data center construction market is booming. In 2021, according to Research and Markets, the market reached $47 billion, and it is expected to reach $75.3 billion by 2027.

To further support this demand, in addition to building new facilities, developers are finding creative ways to adapt existing structures to serve as data centers. The adaptive reuse of data centers has been growing over the last two years, as more facilities have been shuttered and more companies have transitioned to hybrid or fully remote workforces. This has left a variety of properties, especially in urban areas that are ideally suited for data centers, vacant and ripe for reuse.

In honor of International Data Center Day today, we walk through some details and the types of facilities best suited to become data centers through adaptive reuse. And as with all real estate conversations, we start with location …


Investors in data centers have targeted locations in and around large urban areas where high-volume energy, water and communications infrastructure is readily available. Adaptive reuse offers the opportunity to have a data center at the heart of this activity without reinventing the utility footprint, which can be complicated by limited power generation, connectivity and water availability.

Developers need to focus on the expectations for that data center and understand what the IT kit for that site is going to look like in order to take advantage of well-configured properties.

However, urban settings are not the only ideal locations for data centers. There are many excellent sites in the suburbs, as well as uptown. Selecting abandoned properties in any of these environments is an opportunity to boost the local economy by expanding business activities and creating jobs.


In their chosen geographies, developers should be looking at a range of densities to cover a variety of customers. Facilities that feature densities of 100 kilowatts to 150 watts per square foot can still be effective at supporting large data, provided the flooring is structurally sufficient to support the equipment. Determining core technical requirements opens new potential sites and creates many possibilities in the adaptive reuse market.

Ideal buildings to transform into data centers feature clear heights to help disperse the heat generated from computing hardware. The bottom line for investors: What can be constructed most efficiently on a site that has the necessary infrastructure to sustain a long-term data management solution?


Office buildings: Office spaces have become increasingly available for adaptive reuse, and some of the biggest data centers in the world are converted office spaces. However, office spaces often present more challenges because they usually don’t have the clear heights required to cool environments effectively. They also can present space and security challenges.

High-performance computing, whether for high-frequency or new data cloud environments, typically does not fit well within a former office environment, and the required structural modifications to the facility may not make financial sense. The desired data center environment combined with the necessary structural load will dictate whether an office building will work.

Warehouses: Warehouses have traditionally been amenable to data center retrofits. The key to developing a data center through adaptive reuse is understanding the potential facility’s regional footprint. When looking to retrofit an office building or warehouse-type structure, owners must be flexible about design options.

Retail stores: Retail space, particularly in hyperscale retail environments like a mall, tends to be of a lower density and can meet data center requirements better than the office space environment. A large, vacant retail space is a model for hyperscale data center solutions. Many vacant malls are in key locations and could be repurposed to provide a good fit.

Our data needs will only continue to grow. By understanding what location, scale and facility is best suited to fit their needs, data center owners and investors have the opportunity to reap immense benefits from the practice of adaptive reuse.

Rich Keagy

Rich Keagy, PE, LEED, is a senior associate and practice leader at Woolpert who collaborates on regional and global projects out of Woolpert’s Charlotte, N.C., office. A civil engineering graduate from the University of Akron, Keagy has nearly 30 years of senior leadership and consulting experience.