Before an owner ever opens a door for us, we can learn a surprising amount about a data center campus from what is already public. Pulsix, the building-intelligence research program at Zebrix Group, does exactly that: it screens envelope, cooling, and power performance from public data, then hands owners a short list of questions worth verifying on site.
The screen is organized as a connected system graph: roofs and envelope, cooling plant, power distribution, and IT load are nodes, and the links between them show how a problem in one place shows up somewhere else. A hot, aging roof raises cooling load; a stressed cooling plant narrows power headroom. Seeing those links is what turns a pile of measurements into a prioritized plan.
Public data is powerful, but it is not the owner's data. So every figure we publish is labeled observed (seen directly in public sources), modeled (estimated from physics and baselines), or owner-required (only the owner can confirm it). That honesty is the point: the screen tells you where to look, not what you will find.
Our public case study of a West Texas AI data center campus walks through the campus map, a timeline of how the site was built, a performance view, and the opportunities a public-data screen surfaces. Explore the case study.
Pulsix is not affiliated with the campus owner or operator. Every figure in the case study is labeled observed, modeled, or owner-required, and nothing here is an assessment of how the facility is actually run.
For owners and operators, a Pulsix data center evaluation converts the public screen into an owner-verified optimization plan — confirmed with your trend data and a site walk, and scoped by the same team that executes HVAC and mechanical work.
Want this in your inbox monthly? Subscribe to the Zebrix Insights Newsletter, or talk to our team.