Per-MW pricing, regional variance, and cost drivers for owners scoping hyperscale & AI builds.
Salary benchmarks across the 14 mission-critical disciplines.
ATD exam prep should focus on design judgment - not memorizing answers. Alongside the program’s 16 hours of instruction, plan two or three 20–30-minute reviews and one 60–90-minute design exercise each week.
A clear study plan covers four tasks:
The goal is to explain whether a design meets its claimed Tier - and what information is still needed. ATD certifies the designer, not the facility.
ATD Exam Prep: Design Scenario Review Workflow
Confirm your eligibility with Uptime Institute before paying. Its published requirements identify licensed Professional Engineers with design-management responsibilities. A construction-management title alone doesn’t establish eligibility. Send your license status, current role, data center experience, and design responsibilities. If your background falls outside the usual requirements, request written confirmation.[5]
Plan around your selected offering - not a general course description. Confirm the dates, time zone, delivery format, required attendance, exam timing and duration, proctoring requirements, and retake terms. Some official descriptions schedule the proctored exam on the final day, so don’t count on a separate testing week.[6] Block off the full attendance window and assign a qualified deputy to handle RFIs, submittals, and urgent design decisions.
Request the current ATD curriculum, participant materials, assigned exercises, examination instructions, and access to the applicable Tier Standard: Topology. Record each document’s revision, access restrictions, and permitted use during class or the exam. Ask when you’ll receive the materials; enrollment doesn’t necessarily mean immediate access.
If materials arrive only when training starts, use the time beforehand to review prerequisites: electrical one-lines, UPS and generator functions, cooling systems, and controls sequences. If you have limited data center experience, discuss your readiness with an Education Manager or consider Data Center Fundamentals.[1] Keep a question log that names exactly what you need to clarify, such as generator paralleling controls or maintenance-bypass operation.
For extra practice, use only project documents you have permission to use, and anonymize them before sharing. Remove client names, addresses, equipment identifiers, security details, network information, and commercial terms. Confirm permission before uploading course materials or project diagrams to any study platform. Removing a client name alone doesn’t make a document safe to share.
Use the official curriculum to turn the Tier standard into a one-page design review reference[1][2]. Compare required system performance - not equipment counts.
These distinctions come from Tier topology requirements. Tier IV also requires continuous cooling and fault-tolerant IT power architecture[3][7].
Equipment counts or uptime alone don't determine a Tier. Use your reference sheet to check whether a design keeps working during maintenance or failure. Link each requirement to a design decision in your discipline, then note how that decision affects other systems[1][2].
Next, test those requirements against system dependencies.
Follow the course sequence through facility, mechanical, electrical, and ancillary systems, then review common omissions and design examples[1][2].
Draw a dependency map with the IT load at the center. Trace power through switchgear, UPS systems, generators, fuel systems, and distribution. Follow cooling through pumps, piping, valves, and heat rejection. For controls, trace power supplies, communications, and automatic functions. Mark shared components in one color so single points of failure stand out.
For each connection, record event → dependency → impact. Test planned isolation separately from unplanned failure: concurrent maintainability and fault tolerance answer different questions. When electrical equipment is isolated, check that cooling and controls remain available. Another electrical path alone doesn't answer that question.
Record unresolved questions against the applicable course objective.
For each scenario, state the Tier and load assumptions. Trace the paths, test redundancy, check maintenance and fault conditions, review controls, and cite the governing criterion. Does the design still meet its claimed Tier during maintenance or a fault? Record missing evidence and mark each conclusion supported, unsupported, or needs clarification.
Use only approved project examples for practice, with confidential details removed.[5][8]
Hypothetical cooling-maintenance exercise: A design intended for Tier III has two chilled-water pumps, two chillers, and one common control panel. Remove one pump for planned maintenance. Check remaining pump and chiller capacity, isolation valves, electrical supply, controls, and heat-rejection capability against the applicable concurrent-maintainability criterion. If control wiring or valve locations are missing, request the control diagram, sequence of operation, and piping details. Record requiring clarification rather than treating an incomplete diagram as proof of compliance or noncompliance.[3]
Apply this checklist during your weekly analysis block. Each review should produce a written decision.
Set aside short blocks to review terminology and official criteria. Reserve one weekly block for a full scenario review and a one-page decision record covering assumptions, maintenance and fault results, missing evidence, and supporting criteria.[1][2]
During major submittals, startup, or commissioning, keep the brief reviews and move the deeper analysis to a specific calendar slot. Log each unresolved issue with the system, assumed Tier, source, and question. Start with gaps that could change your conclusion, such as unverified remaining cooling capacity or a shared control supply.[2]
As the exam approaches, use your notes to focus on the Tier criteria you know least well.
Work backward from your confirmed exam date. Revisit weak areas and use authorized exercises to practice pacing. Confirm the exam date, format, duration, proctoring, ID, technical requirements, and permitted resources directly with Uptime Institute. Follow the current course instructions.[2]
After each study block, revisit the official standard and your scenario notes. Base your plan on official materials and approved practice. Use Tier criteria to check how power, cooling, and controls work together.[5][4]
ATD certifies the designer, not the facility. Treat it as a design credential - not a replacement for project judgment or facility certification. Apply Tier concepts only when they fit the owner’s objectives, codes, and project criteria.[5][7][9]
You’re ready when you can apply Tier Standard: Topology requirements to actual designs, not just recall them. That means understanding how redundancy, cooling, and distribution paths differ across Tiers I–IV - and spotting gaps in concurrent maintainability or fault tolerance in electrical and mechanical diagrams.
You should also be able to identify noncompliant elements in complex submittals and find technical rules quickly using your course materials and tabbed index, even under time pressure.
Prioritize Tier Standard: Study how topology and design differ across Tier levels, especially concurrent maintainability versus fault tolerance. Watch for pitfalls such as single-corded equipment in dual-path facilities and failures in shared controls.
Practice spotting noncompliant layouts in electrical one-line diagrams and mechanical flow diagrams. Work through design-review scenarios and TCCF templates to apply these concepts - not just memorize them.
Use your internal design review framework to study Tier Standard: Topology and the requirements for each level. Then apply those requirements to sample submittals, electrical one-line diagrams, and mechanical flow diagrams [1].
Review TCCF documents to understand how design choices affect redundancy, maintainability, and fault tolerance [2]. Check for compliance pitfalls that often cause design packages to fail, including single-corded equipment in dual-path facilities and shared control failures [1].
