The ever-increasing digitization of business is forcing IT to reevaluate their current computer power capacity, driving the expansion of existing data centers, or in some cases, the construction of new data centers, including modular data centers. However, today’s competitive marketplace means businesses cannot write a blank check just to meet these increasing demands; IT must approach expansion or new builds with consideration for both capacity and total cost of ownership (TCO).

Whether upgrading an existing data center or building a new one, it’s important to consider TCO. And as one of the more costly data center components to purchase and operate, power and cooling should be a key TCO consideration during the planning phase — both in terms of upfront investment (capital expenses, or CAPEX) and long term spend over the life of the data center, which quickly becomes more expensive than the original investment (operational expenses, or OPEX).

To enable data center owners and operators to maximize TCO, the traditional power and cooling model of customized infrastructure built for maximum capacity is giving way to more modular, engineered, and prefabricated options. Why? Prefabricated modules are pre-assembled, integrated components with modular infrastructure that can scale as the data center load increases over time. Controls are pre-programmed, and the construction process is further aided by existing reference designs. Investment risk is minimized due to lowered costs down the road.

Prefabricated modules also simplify both greenfield builds with no existing infrastructure and brownfield builds where an existing site is being repurposed. Power and cooling is scaled and right-sized to actual data center loads, removing the need for guesswork and estimations that can result in the purchase and operation of more infrastructure than actually necessary.

According to a 2014 Schneider Electric white paper, in an ideal situation, prefabricated infrastructure can yield TCO savings of nearly 30% over a traditional data center — with an average of 27.2% CAPEX savings and 31.6% OPEX savings.


The standardization and predictability of pre-assembled, integrated modules save deployment time and upfront costs when compared to the same electrical and mechanical infrastructure implemented with custom engineering and considerable onsite work. However, the savings don’t end there. The modular nature of prefabricated power and cooling infrastructure enables scaling and right-sizing to actual data center loads, which is responsible for a large portion of potential savings.

Right-sizing infrastructure. The single largest avoidable cost associated with traditional data center builds is oversizing. When considering a 10-year operational cost breakout of a traditional 5 megawatt (MW) data center at just over $52 million, vs. a modular 4 MW data center at just over $35 million, avoiding overbuilt capacity accounts for approximately 12% of potential OPEX savings.

This is because traditional data center design is planned for worst case final load to avoid penalties for running out of capacity during the data center’s lifecycle. However, final load rarely actually hits this number. Prefabricated data centers, on the other hand, help avoid overbuilt capacity because modules can be purchased on an as-needed basis, and easily added to when there is a real need for more capacity.

Scalability. Because prefabricated modules are built to scale over time, capital and maintenance costs are deferred until they’re actually needed to support the increased load demand. Therefore, systems run at an overall higher percent load each year, resulting in energy savings. Contrast this to traditional data center builds where small loads relative to overall capacity results in an extremely low power usage effectiveness (PUE). Right-sizing can result in dramatic savings, potentially eliminating up to 50% of the electrical bill. This is why, when using the same model described above comparing a 5 MW traditional build to a 4 MW modular build, the ability to scale over time accounts for almost 6% of OPEX savings.

Additional savings. Additional savings can also be realized by implementing prefabricated, modular infrastructure. For example, row cooling, or close coupled cooling, reduces energy cost because it shortens the path to servers. Factory assembly with integrated components and controls are less expensive than building from a collection of multiple vendor parts in the field. Also, valuable time spent calibrating cooling system controls through the integration of fans, pumps, loops, chillers, cooling towers, and more are reduced.

Prefabricated modules are more compact, reducing the amount of space required to cool, heat, and ventilate the area. This smaller core and shell can result in savings of $100 to $150 per square foot. The influence of other systems that are often contained within traditional facilities, such as building comfort cooling systems, is reduced or non-existent, reducing overcooling (or cooling a larger space than necessary). Integrated cooling controls also provide more predictable and reliable cooling.


These TCO savings can be applied to both new data center builds and existing data center expansions. However, there are also benefits specific to new builds; prefabricated modular data centers can speed up planning and deployment time, resulting in significant cost savings.

Design/planning. For example, during the planning and design phase, engineering and specification is performed at the system level, vs. the component level for traditional builds. The overall timeline for this phase is typically less than 12 weeks, whereas traditional data centers can take up to 24 or more weeks.

Site preparation. When considering site preparation, modular data center components can be placed either indoors or outdoors. For traditional builds, placement can only be indoors (with the exception of heat rejection and generators). Modular power and cooling infrastructure can also be placed directly over underground wiring and piping, vs. the need for additional (and sometimes costly) support hardware or special structures for traditional data centers.

Procurement. During the procurement phase, modules are typically designed to meet transportation weight limits and arrive on site nearly fully assembled. Traditional data center components are often shipped separately from multiple vendors, which can cause project delays and potential confusion as to when and where components arrive. Modules also minimize construction trash due their more compact shipping state.

Installation. Prefabricated modules are typically installed by truck crane, whereas traditional data centers require separate products to be lifted by forklifts or by hand — adding both time and the need for additional manual labor.


While prefabricated power and cooling modules offer benefits in terms of the ability to right-size infrastructure, increase scalability and more, there are other methods data center owners and operators can implement to increase TCO savings even more.

Reference designs. Reference designs are tested, validated, and documented design packages for physical data center infrastructure systems. Reference designs can aid in the design process for entirely new data center builds, and there are also reference designs for specific subsets — for example, a power or cooling plant, or an IT room. When combined with prefabricated power and cooling modules, owners and operators are able to visualize and put into practice a comprehensive, cost-effective blueprint for their expansion or new facility — removing much of the guesswork and virtually eliminating unforeseen complications or obstacles.

Reference designs outline the bill of materials, electrical and mechanical diagrams, and floor layouts. They also include system-level performance attributes based on actual prior experience, so owners and operators can understand and make decisions based on the expected capacity. Not only does this reduce time spent calculating projected performance based on individual components, it’s also a more accurate method for doing so. Owners and operators can compare designs and quickly uncover potential tradeoffs; this process would typically take weeks or even months as opposed to mere minutes or hours.

Reference designs also:

• Simplify the planning phase because they have been tested and validated, reducing the amount of time spent on those tasks

• Reduce risk because system performance is guaranteed by the design

• Reflect collective intelligence and lessons learned from previous installations, resulting in a more accurate and reliable design

By combining the benefits of prefabricated data center modules with reference designs, owners and operators can maximize TCO and reduce CAPEX and OPEX expenditures. 

Data center infrastructure management (DCIM) tools. To maximize the benefits provided by prefabricated power and cooling modules, owners and operators should consider implementing data center infrastructure management (DCIM). DCIM solutions are able to aggregate data from siloed systems to provide an integrated view of the physical building structure and IT equipment from the building down to the server level, enabling owners and operators to further improve planning and cut operational costs. 

A comprehensive DCIM solution can provide the following functions:

• Visibility into the status and configuration of physical systems

• Notification of status changes

• Remote configuration of facility power, cooling, and security system settings

• Tracking of data center assets

• Facilitation in the deployment of new equipment

• Facilitation in the planning of changes to the data center

• Simulation of potential changes to analyze potential impact With these functions, DCIM can aid in the following processes:

Planning. A DCIM solution shows the current physical state of the data center and simulates how adding, moving, or changing future physical equipment can affect operations. Owners and operators can then use this information to answer common planning questions.

• Operations. DCIM allows owners and operators to assign work orders, reserve space, track status and extract an audit trail, providing visibility into and history of the change cycle whenever equipment goes in or out of the data center.

Analysis. Owners and operators can use information derived from the DCIM solution to make optimal, realistic decisions. It can also determine the cause of problems, as well as actually generate business value. Established baselines created by the collection of data over time can be established to uncover opportunities for increasing efficiency and reduce overall costs.

These functions greatly reduce the gap between facilities and the data center, which leads to smarter, more effective decision-making that betters the health and efficiency of overall systems  — thereby optimizing CAPEX and OPEX, and mitigating operating risk down the road. When used in conjunction with prefabricated data center power and cooling modules, as well as reference designs, owners and operators can rest assured that their new data center build or expansion goes as smoothly and predictably as possible — while getting the best bang for their buck in terms of TCO.


It’s important to remember the primary benefits to prefabricated data center modules over traditional data center designs. Traditional data center designs incorporate excess capacity upfront to account for potential expansion, making it difficult to expand power and cooling capacity later on. Designs are overly conservative in capacity planning, resulting in higher upfront capital costs and greater inefficiency.

Prefabricated modules eliminate wasteful oversizing because of its module architecture that makes it easier to add or reduce capacity to meet real-world demand. In conjunction with efficient, integrated power and cooling technology, prefabricated modules can result in up to 30% TCO savings compared to the traditional data center. By also taking advantage of reference designs and DCIM tools, these savings can be even greater.