Methodology for Implementing Modular Data Centers and Interpreting Standardization Processes

Personal disclaimer: the images are publicly available or created by the author; the discussion reflects the author's professional perspective and does not represent any company's views.

Key questions: definition of modularity, how to judge it, who drives modular data center adoption, the real benefits of rapid deployment and lower PUE, the rise of micro‑modules in China, and guidance on selecting and implementing modular solutions.

7 Implementation Methodology for Modular Data Centers

Before starting, the author poses several industry‑level questions about the roles of consulting design firms, equipment suppliers, and integration solution providers, the differing modular designs of Google, Microsoft, Facebook, and Yahoo, how to assess modular depth, and where to begin a modularization effort.

The answer introduces the crucial concept of standardization. Without repeatable, copy‑able modules, modularity has no practical value. Therefore, modular data centers are essentially the result of a standardization process that spans design and construction phases.

Standardization design is likened to a doctor's prescription, while standardization prefabrication corresponds to producing the medicine based on that prescription. Both are required for true modularity.

Using China's “micro‑module” as an example, the author shows that it typically achieves only standardization design, not full prefabrication, unless additional on‑site prefabrication steps are added.

Containerized data centers, by contrast, represent full standardization prefabrication, with components assembled in a factory and shipped as a complete unit.

Figure 16 – Example IT container.

By classifying data‑center equipment into ICT and facility categories and mapping them onto the two layers of standardization, a matrix of the standardization process can be derived.

Figure 17 – GE Power Mod.

Figure 18 – Standardization process diagram.

Thus, modularity is the outcome of standardization.

8 Interpreting the Data‑Center Standardization Process

The remaining three questions from Chapter 7 are answered by applying the standardization framework to leading internet companies. Google’s timeline (image) shows a high degree of standardization across both design and prefabrication. Similar analyses for Microsoft and Facebook reveal comparable levels of standardization.

For colocation providers such as Digital Realty Trust, the standardization level is lower because they lack control over ICT customization. Their focus is on mechanical‑electrical modules and building modules, where higher standardization yields greater modularity benefits.

Figure 21 – Digital Realty Trust images.

The discussion concludes that the degree of standardization varies by business model, with internet companies achieving higher levels due to in‑house ICT development, while pure colocation services prioritize flexibility over full modularity.

(To be continued…)