Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more

Plan for the exact AI servers and workload you intend to install, not a generic “high-density” rack rating. First establish the equipment’s electrical and thermal requirements; then confirm utility and facility capacity, design the power paths and redundancy, choose cooling that fits the hardware and site, and commission monitoring and failure response. A rack-density figure by itself cannot show whether a facility can deliver the required power or remove the heat.

What information do you need before sizing a rack?

Define the workload, equipment, and deployment phases

Record the AI workload, selected server platform, rack configuration, deployment schedule, and availability and maintenance requirements. Obtain the manufacturer’s electrical and thermal specifications for that exact configuration, including peak demand and any operating assumptions. A published design for one server generation or cluster is not a sizing rule for another.

For example, Schneider Electric’s September 2026 Reference Design 113 is scoped to Vera Rubin NVL72 clusters. Its March 2026 Reference Design 100 describes separate air-cooled and liquid-cooled retrofit cases. These examples illustrate how designs vary by equipment and site; neither should be treated as a universal rack specification.

Free tools Windows power users keep installed

One-click scans. No signup required.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Set the design boundary

Decide whether the project covers a rack, row, pod, or a larger facility expansion. Record the load and required availability at each boundary, because a rack that fits the room’s cooling capacity may still exceed a circuit, UPS, distribution panel, or utility limit.

#1 Best Overall
Sale
ZHANGONG 19in 1U Metered PDU Rack Mount Power Strip with Display 7 Outlets
  • 【360° Rotating Mounting Brackets】 – Innovative 360-degree rotating brackets enable flexible installation on server racks, walls, desks, and furniture. This 1U PDU mounts horizontally in any standard 19-inch server rack, delivering versatile power distribution for data centers, network cabinets, and audio-visual setups.
  • 【7-Outlet 1U Rackmount PDU】 – Designed for standard 19-inch server racks, this 1U rack mount power strip provides 7 outlets with 15A/125V/1875W capacity. Ideal for powering servers, switches, routers, and networking equipment in professional IT environments.
  • 【Real-Time Power Monitoring & Overload Protection】 – Built-in digital display provides real-time voltage and current readings for precise power management. The 15A overload protection switch automatically trips when current exceeds safe limits—circuit breaker responds only to overload conditions, not accidental wire contact, ensuring reliable, uninterrupted operation.
  • 【Heavy-Duty Metal Housing & 14AWG Cord】 – Constructed with rugged, industrial-grade metal housing for maximum durability in demanding environments. Features a 14AWG heavy-duty 6.5FT power cord delivering stable, consistent power delivery for high-demand server rooms and industrial settings.
  • 【1U Rackmount with Cable Management】 – Occupies just 1U of rack space, maximizing valuable server rack real estate. Includes mounting brackets for horizontal installation in standard 19-inch racks, with a streamlined design that simplifies cable management and maintains a clean, organized network cabinet.

Can the site supply the power and cooling?

Check capacity and schedule before committing to a layout

Assess utility service and expansion plans, interconnection timing, transformer and switchgear availability, distribution capacity, backup-power strategy, permits, cooling resources, water constraints, and space for phased growth. ASHRAE’s AI Data Center Energy Performance Framework site-planning guidance says to “Prioritize power availability and grid capacity early in site selection, as utility constraints and interconnection timelines are now primary drivers of feasibility and schedule risk.” Treat utility coordination and equipment lead times as feasibility and schedule items, not late-stage procurement details.

Evaluate the whole site, not only watts per area

Area power density does not establish whether the site can serve a particular high-density load. Confirm that the power path, heat-removal capacity, and expansion plan work for the proposed equipment and its location. Include climate and hazards, water availability, workload fit, and phased-growth needs in site selection. ASHRAE points planners to its TC 9.9 thermal guidance for recommended and allowable IT environmental envelopes.

Rank #2
Sale
Tripp Lite Metered PDU, 20A, 12 Outlets (5-15/20R), 120V, L5-20P / 5-20P, 110-127V Input, 15 ft. Cord, 1U Rack-Mount Power, 2 Year Warranty (PDUMH20) Black
  • 20a single phase 120V Metered power distribution unit/ PDU (agency de-rated to 16a continuous)
  • Built-in 2 digit Visual meter continuously reports PDU load level in Amps
  • Attached NEMA L5-20P 20a 120V input Plug with 15 ft./ 4.5m line cord
  • 12 NEMA 5-15/20R outlets
  • 2-Year limited warranty

How should you size rack power distribution?

Trace demand through each electrical boundary

Use the selected server’s documentation and a qualified electrical design to calculate demand at the rack, row or pod, UPS, and facility boundaries. Document the actual distribution voltage and phase, circuit protection, applicable breaker derating, power-factor assumptions, expected peak loads, and cable, plug, and outlet requirements. Check the requirements against local electrical codes and the equipment documentation.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

NVIDIA’s DGX H100 SuperPOD guide gives a platform-specific example: for its listed 415 V, 32 A, three-phase Wye case, the table states 21.8 kW circuit capacity and 20.4 kW peak server demand per circuit. Its footnotes specify a 0.95 power factor and the guide’s N+1 provisioning assumptions. The guide says its preferred power for high-density deployment patterns is “415 VAC, 32A, three-phase, N+1.” Those values and that preference apply to the DGX H100 design patterns described in the guide, not to other platforms or sites. It also says more than four DGX H100 systems per rack is not recommended for thermodynamic reasons.

Rank #3
StarTech 8-Outlet 1U Metered PDU, Surge, TAA (8NM8-RACK-MOUNT-PDU)
  • 8-OUTLET 1U RACK MOUNTED METERED PDU: 8x NEMA 5-15R power outlets; 125V max and 15A circuit breaker; Built-in 8ft (2.4m) NEMA 5-15P power cord; Ideal power supply strip for 19" racks in server rooms/closets, data center, and testing labs
  • METERED POWER DISTRIBUTION UNIT: Large-font display with optional backlight for real-time monitoring of key power parameters: voltage (V), current (A), wattage (W), and power factor (PF), which indicates the efficiency of the connected devices
  • SURGE PROTECTOR: Self-tripping MOV design with mechanical trigger instantly cuts power during surges to protect connected devices; Includes LED status indicator, resettable 15A circuit breaker, and safety cover to prevent accidental resets
  • RACK MOUNTABLE: 1U horizontal rackmount PDU for EIA-310-compliant 19-inch server racks/cabinets that are 4in (10.1cm) or deeper; Made of cold-rolled steel w/textured finish for sturdiness & long-lasting durability; 6x rear and 2x front electrical outlets

Use metered rack distribution where it supports operations

A metered rack PDU can provide rack-level power visibility, but its suitability depends on the installation. Match voltage, phase, current rating, plug and outlet configuration, redundancy, rack fit, and monitoring capability to the server and facility design. Specify qualified installation; do not choose a unit based on a generic marketplace description.

Which redundancy arrangement matches the failure you must withstand?

Map sources and shared points of failure

Start by listing the events the design must tolerate: loss of a rack circuit or PDU, a floor PDU or remote power panel, a UPS, a utility source, or an equipment-maintenance event. Draw the path from each source to each rack and identify any points where apparently separate feeds share an upstream device. A label such as “N+1” does not, by itself, show what remains powered after a failure.

Rank #4
StarTech 12-Outlet 1U Metered PDU, Surge, TAA (12NM8-RACK-MOUNT-PDU)
  • 12-OUTLET 1U RACK MOUNTED METERED PDU: 12x NEMA 5-15R power outlets; 125V max and 15A circuit breaker; Built-in 8ft (2.4m) NEMA 5-15P power cord; Ideal power supply strip for 19" racks in server rooms/closets, data center, and testing labs
  • METERED POWER DISTRIBUTION UNIT: Large-font display with optional backlight for real-time monitoring of key power parameters: voltage (V), current (A), wattage (W), and power factor (PF), which indicates the efficiency of the connected devices
  • SURGE PROTECTOR: Self-tripping MOV design with mechanical trigger instantly cuts power during surges to protect connected devices; Includes LED status indicator, resettable 15A circuit breaker, and safety cover to prevent accidental resets
  • RACK MOUNTABLE: 1U horizontal rackmount PDU for EIA-310-compliant 19-inch server racks/cabinets that are 4in (10.1cm) or deeper; Made of cold-rolled steel w/textured finish for sturdiness & long-lasting durability; 10x rear and 2x front electrical outlets

Compare arrangements by outcome, not label

Arrangement What to establish Trade-off to evaluate
Basic or enhanced N+1 Which components or paths have spare capacity, which loads remain energized after a fault, and whether the paths share upstream equipment. Fault tolerance, maintainability, and the number and independence of upstream paths. NVIDIA describes basic and enhanced N+1 schemes for its DGX H100 architecture; its more independent arrangement may require three discrete upstream paths.
2N Whether the intended two paths are genuinely independent through the relevant distribution chain, and how the server connects to them. Resilience, equipment compatibility, capital cost, and maintenance requirements. Vertiv’s 2024 reference design illustrates a separate 2N arrangement.

Compare the paths against the required failure and maintenance scenarios. Do not assume different redundancy labels are interchangeable or that one is best for every facility.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Support on Ko-Fi

Which cooling topology fits the servers and facility?

Compare air and liquid approaches against site conditions

Approach When to evaluate it Questions to resolve
Air cooling Where the equipment and room design can support the load with the existing or planned air-cooling system. Can the system remove the servers’ heat at the planned load? What room cooling, electrical changes, and expansion capacity are needed?
Liquid cooling with a liquid-to-air CDU Schneider Electric’s Reference Design 100 presents this as a retrofit option for sites without facility-water systems. How will the CDU reject heat to air, and can the surrounding facility support the required heat rejection, residual room heat, power, and controls?
Liquid cooling with a liquid-to-liquid CDU Schneider Electric’s Reference Design 100 presents this where facility-water systems are available. Is the facility-water system compatible with the design, and how will the CDU, facility loop, heat rejection, electrical systems, and controls work together?

For any option, verify the server’s supported cooling method and heat load, including heat that remains in the room. Compare water availability and constraints, local climate and ambient conditions, existing infrastructure, retrofit limits, redundancy, maintainability, monitoring, energy and water implications, equipment lead times, and expansion plans. A CDU is part of a facility cooling topology, not a stand-alone accessory recommendation.

Best Value
Sale
StarTech 8-Outlet 1U PDU, 120V/15A, Surge, 6ft Cord, TAA (RKPW081915)
  • POWER AND CHARGE: This rack mount power strip provides an additional 8 NEMA 5-15 outlets (120V/15A) and features a 6ft (1,8m) long cord so you can plug your devices in while leaving the rack mobile
  • 1U RACK DESIGN: Compatible with all 19" server racks 4 inches or deeper, this horizontal-mount power distribution unit fits many network racks and has an integrated power cord; ANSI/EIA RS-310-D standard
  • EASY INSTALLATION: This IT-grade rackmount PDU features a rugged steel chassis, LED indicators for ground and surge protection, and lets you control the power state with power and reset switches
  • PROTECTS YOUR EQUIPMENT: This rack mountable 8-outlet (120V) power strip features a built-in circuit breaker and reset switch, ensuring a dependable performance of your networking equipment
  • THE IT PRO'S CHOICE: Designed and built for IT Professionals, this rack PDU is backed for 2-Years, including free lifetime 24/5 multi-lingual technical assistance

Interpret reference-design ratings narrowly

Schneider Electric’s 2026 Reference Design 100 is titled as a 3,818 kW design and is scoped to a North American Tier III chilled-water design covering air-cooled and liquid-cooled AI clusters. That title rating is for the particular reference design, not a per-rack requirement or general industry limit.

What should you verify during commissioning and operation?

Check the installed power paths before powering servers

  • Confirm that feeds originate from the intended distribution paths and that circuit and kVA ratings match the approved design.
  • Verify that labels identify the power sources, distribution points, and breakers, and confirm protective grounding.
  • Test the specified redundancy arrangement, including whether the intended loads remain served during the planned breaker-by-breaker or other fault checks.
  • Validate phase balance and the installed feed configuration against the equipment and electrical design.

NVIDIA’s DGX H100 guide calls for qualified facilities personnel to verify supplied capacity and describes breaker-by-breaker AC verification of rack redundancy. Apply verification procedures appropriate to the selected equipment and local electrical codes.

Maintain visibility after handover

Provide rack-level power monitoring and environmental visibility so operators can see conditions relevant to the design. NVIDIA’s guide recommends remote rPDU monitoring with API capability and rack temperature and humidity monitoring. Define who receives alerts, how operators respond to a failed feed or cooling component, and how monitoring is checked during maintenance.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

How should you use vendor reference designs?

Use them to identify design questions and compare approaches, not as substitutes for a site-specific engineering design. ASHRAE provides broad site-planning guidance; NVIDIA’s electrical example is for DGX H100 SuperPOD; Schneider Electric and Vertiv illustrate vendor-specific reference scenarios. Confirm every proposed configuration against the selected equipment, utility and facility conditions, applicable codes, and qualified engineering review.

Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.