The expectations and requirements of data center customers will continue to shift with new technologies and evolving security considerations. Following those shifting requirements are innovations in power distribution, cooling, and monitoring technologies. As data center managers work to build data centers of the future, the scalability of equipment including cooling, power, and protection will continue to be a key consideration.
There are more than 5,000 data centers in the US, and data infrastructure demand is increasing. However, the workload demands of artificial intelligence, machine learning, and high-performance computing cannot be met with new construction alone. The availability of real estate, labor constraints, and the simple time it takes to construct new facilities will all require data centers to find creative solutions to meet customer needs.
Data centers must be able to scale within their existing facilities and build new facilities with scalability in mind.
Rack-Level Modularity is Key
Modular data center equipment allows data centers to scale operations as demand increases. If data centers build to the capabilities required today, they will quickly find themselves conducting rework and further developments as technology changes. Data center managers need to preserve the ability to add additional racks or equipment within existing building infrastructure to scale with rapidly increasing demand that cannot be met in time by larger expansion projects.
Cooling technology has the potential to be a large driver of modularity. Cooling systems for different kinds of equipment – for instance, high-density liquid cooling for high-performance chips and hybrid solutions for more standard IT – need to have the ability to be flexed in and out to fit deployed equipment.
As cooling technology continues to improve, data centers can also drive greater density of IT. However, this means cabling and power distribution must then also be designed with a scalable architecture in mind.
Data center managers need to constantly manage and monitor their infrastructure to assess where and when to upgrade systems. Users of data centers should be able to monitor power, cooling, and other capabilities remotely and in real time to ensure those systems meet their requirements.
Modularity as a Component of the Data Center Infrastructure Evolution
Innovation in the IT and data center industry advances rapidly, particularly with the paradigm shift of AI rollout. Data center operators face uncertainty regarding the future balance between air-cooled and liquid-cooled deployments, not only in the coming year but also over the next 5 or 10 years.
This uncertainty is compounded by evolving IT system architectures and chip innovations, which makes it challenging to predict future cooling and power infrastructure requirements. Modularity therefore has become a crucial factor in avoiding overcapitalization on data center power and cooling infrastructure, providing flexibility in adapting to changing needs.
Modularity for Redundancy
Modularity has additional benefits outside of just scalability. Using prefabricated and interchangeable technologies can help data centers maintain uptime and performance since critical components can be exchanged when service is necessary.
When identifying enclosures, power distribution, cable management and liquid cooling solutions to deploy in data centers, data center managers need to make sure they are designed with scalability and modularity in mind.
For instance, cooling units should be able to be connected to either coolant distribution units or to facility water supplies, if such infrastructure exists. This allows data center managers to use the same manifolds at the rack level whether they are using liquid-to-air technologies, cooling distribution units, facility liquid lines, or any combination of systems.
Scalability at Every Level
Facility engineers designing, retrofitting, or remodeling buildings for data center occupancy must also keep scalability in mind. This might mean installing more infrastructure to support advanced cooling capabilities such as liquid lines to racks and rows. It can also mean thinking about future energy requirements by budgeting for on-site renewable energy and energy storage. This can offset energy use, allowing data centers to grow in density without expanding physical footprint.
Matt Archibald is the Director of Technical Architecture at nVent, supporting the data center and networking space. Matt is focused on liquid cooling. unified infrastructure management, data center monitoring, and automated data center infrastructure management.
