Data Center Infrastructure
CSI Division 27, Communications
The specialized structured cabling, cable pathways, and raised access flooring that carry and organize a data center's network connections, paired with the redundant power distribution, precision cooling, and specialized fire suppression built around it, together engineered to a specific reliability tier, the Uptime Institute's Tier I through Tier IV classification, for example, based on how much downtime the owner can tolerate. The communications and cabling scope sits in this CSI division; the redundant power, cooling, and fire suppression systems it depends on are designed under their own respective divisions.
Why it matters
A higher reliability tier requires genuinely redundant infrastructure, such as two completely independent electrical and cooling paths, not just backup equipment sitting idle, and building to a tier the owner's actual uptime requirements don't justify wastes significant construction budget on redundancy the owner doesn't need.
On a real project
A financial services company specifies its data center be built to Tier III reliability, requiring concurrently maintainable redundant power and cooling paths so any single piece of infrastructure can be taken offline for maintenance without interrupting server operations.
Who this matters most to
A MEP Engineer designs the redundant power and cooling infrastructure to the specified tier. A Commissioning Agent verifies the redundancy actually functions as designed, including testing that a single failure doesn't take down the whole facility.
Where this goes wrong
A data center's electrical infrastructure is designed with two power paths on paper, but both paths are found during commissioning to share a single upstream transformer that was never actually made redundant. The facility doesn't discover the shared single point of failure until a commissioning test deliberately fails one path and the whole facility loses power instead of failing over cleanly.
Explore next