Texas Data Center: Cross-Building Generator Exhaust Exterior CFD
Three adjacent buildings, one shared flow field — the cohort's first campus-scale case. Public-source inputs. Screening-level.
What This Page Covers
This page presents a screening-level, campus-scale exterior CFD analysis of three adjacent data centers on a 42-acre campus in Garland, Texas, designed by a national A+E firm with a dedicated mission-critical practice for full architecture and engineering delivery, including MEP, fire protection, sustainability, ICT, and interior design. The first building on the campus was recognized by industry press for its architectural design.
This is the cohort's first multi-building case. The three buildings sit side by side — Building A on the west, Building B in the middle, Building C on the east — so each building's heat rejection discharges into a flow field shaped by its neighbors. The model applies active rooftop chiller heat rejection on Buildings A and B. Building C is included as passive geometry only: its rooftop equipment could not be confidently identified from public imagery, and consistent with the survey's public-source discipline, we model what we can verify rather than guess at exhaust characteristics. Building C's mass still shapes the campus flow field as an obstruction.
Facility Context
The facts below are derived entirely from cited public sources: the operator's published facility specifications, the A+E firm's portfolio attribution, facility tour coverage in industry press, and satellite imagery. No proprietary drawings, specifications, or operational data are used.
Location: Ashburn, VA.
Owner-operator: A major northern Virginia colocation operator.
Critical IT capacity: 16 MW across six vaults.
Data floor: ~112,000 sq.ft.; two-story building; opened March 2018.
Cooling: Waterless pumped-refrigerant cooling, N+2 redundancy. Rooftop array; no cooling towers, no chilled-water loop. The CFD treats the rooftop heat-rejection units in chiller-style abstraction for the purpose of intake-temperature analysis.
Backup power: 22 diesel generators in two pods of 11 ("11 to make 10" topology, as observed in satellite imagery). Ground-level generator yard along the south building face.
Electrical service: 34.5 kV distribution, two redundant utility feeds.
Source attribution: National A+E firm with a dedicated mission-critical practice.
Scenario Presented
Wind and ambient.
Wind from the east (90°) at 10 mph, 105°F ambient. The easterly direction is the case that aligns generator exhaust transport with the rooftop chiller arrays.
Operating scenario.
All generators running at full load with rooftop cooling at N (half of the installed 2N chiller capacity) on both Buildings A and B — a realistic on-generator operating state rather than a maximum-everything condition. Building C passive.
This page presents the easterly case. Additional wind directions — including north and south, which raise distinct chiller-to-chiller interaction questions — were examined as part of the directional screening and are available to the engineer of record on request.
What the Model Shows
The observations below are qualitative readings of the presented scenario. Quantitative outputs — per-unit intake temperatures, recirculation magnitudes — are shared with the engineer of record on request.
Generator exhaust impinges on the rooftop chillers. Under the easterly wind, flow patterns develop that carry generator exhaust onto the rooftop chiller arrays — a condition that can drive chiller intake temperatures toward the ~125°F regime where shutdown setpoints engage on common air-cooled platforms. This is the same generator-to-chiller pathway seen elsewhere in the cohort, here at campus scale.
The neighboring building shows the same exposure. The phenomenon is not confined to one roof: the easterly case produces comparable generator-exhaust impingement on the adjacent building's chiller array. A campus model surfaces this; a single-building model would miss it.
North and south winds raise a chiller-to-chiller question. With the buildings sitting side by side, winds along the campus axis can carry one building's chiller discharge toward its neighbor's intakes. The directional screening flags this as a potential interference mode worth dedicated cases — detail available on request.
Building geometry governs the result. The three-building arrangement — including the passive Building C mass — shapes the campus flow field, sheltering some arrays and channeling flow toward others depending on wind direction. The exposure is a property of the campus layout, not any single rooftop in isolation.
Methodology
The methodology applied here is the same standardized exterior CFD approach applied to every facility in the cohort — cylindrical far-field domain, logarithmic atmospheric boundary layer inlet, polyhedral mesh in Siemens STAR-CCM+, realizable k-ε RANS baseline. The presented case uses a 105°F ambient, at the upper end of the Dallas–Fort Worth summer design range (site elevation ~538 ft). Full domain setup, boundary conditions, solver choices, and stated limitations are documented at the Methodology page. Key terms used here are defined at the Key Terms and FAQ.
For the Engineer of Record
The full campus figure set, the complete wind-case matrix, and a quantitative summary are available on request. We share these directly with the named engineer of record, not with building owners, operators, or other parties. Natural extensions include activating Building C with verified equipment data, additional wind and seasonal cases, and a forward-looking parametric study of how the campus thermal envelope shifts as chip-level liquid cooling absorbs a growing share of IT heat. Contact stewart@resolvedanalytics.com and reference this Garland, TX campus.
Abut the Author
Stewart Bible, Principal, Resolved Analytics. Resolved Analytics is a Computational Fluid Dynamics consulting practice and authorized Siemens STAR-CCM+ reseller, with a long-standing service line in mission-critical facility exterior analysis. Contact: stewart@resolvedanalytics.com.
Disclosure
This is independent research conducted by Resolved Analytics without engagement, sponsorship, or input from the building owner, operator, or engineer of record. All inputs are derived from cited public sources; no proprietary drawings, specifications, or operational data are used. Results represent idealized exterior conditions and do not represent the actual as-built performance of any facility. No claims are made regarding life-safety, code compliance, or operational performance. All firm and project references have been anonymized. This material is not engineering services rendered to any party.
Campus-scale exterior CFD output: wind from the east (90°) at 10 mph, 105°F ambient, all generators running with cooling at half of installed capacity (N of 2N). Surface coloring on the rooftop chillers shows local intake temperature; Building C is included as passive geometry.
Plan view. Chiller intake temperatures with streamlines under the easterly case. Cross-building transport appears where one roof's discharge carries onto the neighboring building's array rather than dispersing downwind.
Elevation view. Same case, side elevation. Plume rise above the roofs and the sheltering and channeling influence of adjacent building masses are visible from this angle.