Infrastructure Planning ToolJan 2026

AI Infrastructure
Power Calculator

Calculate accurate power consumption, cooling loads, electrical infrastructure requirements, and operating costs for your AI GPU deployment. From single workstations to multi-rack data center configurations, plan your facility with confidence.

20+ GPU Models
72W-2.7kW TDP
30-40% TCO
Real-Time Calc
GPU Power
TDP × Utilization × Count
Cooling
BTU/hr + Tons Required
Electrical
Amps + Circuit Sizing

Executive Summary

Unprecedented Power Density

Modern AI GPUs consume 700W-1,100W each. An 8-GPU server can draw 10kW or more, creating facility challenges that traditional IT infrastructure never faced.

Significant Operating Costs

Electricity represents 30-40% of total AI infrastructure cost over 5 years. Accurate planning prevents budget overruns and identifies optimization opportunities.

Infrastructure Requirements

GPU deployments require proper electrical circuits, cooling capacity, and rack power distribution. Underestimating leads to costly retrofits or capacity constraints.

Plan Before You Deploy

Use this calculator to determine power, cooling, and electrical requirements before procurement. Proper planning ensures smooth deployment and optimal operations.

Understanding AI Infrastructure Power

GPU Thermal Design Power (TDP)

TDP represents the maximum sustained power a GPU can draw under typical workloads. Modern AI accelerators have dramatically higher TDP than consumer GPUs:

  • Consumer GPUs: 200W-575W (RTX 4090: 450W, RTX 5090: 575W)
  • Workstation GPUs: 72W-350W (L4: 72W, RTX 6000 Pro: 350W)
  • Datacenter GPUs: 350W-1,400W (H100: 700W, B200: 1,000W, B300: 1,400W)
  • Superchips: 2,700W+ (GB200 NVL72 per tray)

Power Usage Effectiveness (PUE)

PUE measures total facility power divided by IT equipment power. It accounts for cooling, lighting, power distribution losses, and other overhead:

Excellent (Hyperscale) 1.1 - 1.2
Good (Modern Colo) 1.2 - 1.4
Average (Traditional DC) 1.4 - 1.6
Poor (Older Facilities) 1.6 - 2.0+

Utilization Impact

GPUs rarely run at 100% TDP continuously. Actual power consumption varies by workload:

  • Idle: 20-40% of TDP
  • Inference (Light): 50-70% of TDP
  • Inference (Heavy): 70-85% of TDP
  • Training: 85-100% of TDP

Plan for 85% average utilization for production inference workloads. Training clusters may sustain 95%+ utilization.

Conversion Factors

Essential conversions for power planning:

Watts to BTU/hr × 3.412
BTU/hr to Cooling Tons ÷ 12,000
kW to Annual kWh × 8,760
Watts to Amps (208V) ÷ 208
NEC Circuit Sizing × 1.25 (80% rule)

Power Requirements Calculator

Configure your GPU deployment and get instant power, cooling, and cost estimates

Configuration

50% (Light) 100% (Max)

Results

Power Consumption

Selected GPU TDP 700 W
Total GPU Power 4.76 kW
Total Facility Power (with PUE) 7.14 kW

Annual Operating Costs

Annual Energy Consumption 62,546 kWh/year
Annual Electricity Cost ¤6,255
5-Year Electricity Cost ¤31,273

Cooling Requirements

Heat Generation 16,241 BTU/hr
Cooling Capacity Needed 1.4 tons

Electrical Requirements

Current Draw (208V) 34.3 A
Current Draw (240V) 29.8 A
Circuit Size (NEC 80% rule) 43 A (208V) / 37 A (240V)
Servers Required 1

GPU Power Specifications

Complete power specifications for AI accelerators from NVIDIA, AMD, and Intel

GPU Power Specifications Reference Table
GPU Model TDP (W) Memory Category Annual Cost*
NVIDIA Consumer
RTX 4090 450W 24GB GDDR6X Consumer $335/year
RTX 5090 575W 32GB GDDR7 Consumer $428/year
RTX 5080 360W 16GB GDDR7 Consumer $268/year
RTX 5070 Ti 300W 16GB GDDR7 Consumer $223/year
RTX 5070 250W 12GB GDDR7 Consumer $186/year
NVIDIA Workstation & Edge
RTX 6000 Ada 300W 48GB GDDR6 Workstation $223/year
L4 72W 24GB GDDR6 Edge/Inference $54/year
RTX 6000 Blackwell Pro 350W 96GB GDDR7 Workstation/Datacenter $260/year
NVIDIA Datacenter (Ampere)
A100 40GB 400W 40GB HBM2e Datacenter $298/year
A100 80GB 400W 80GB HBM2e Datacenter $298/year
NVIDIA Datacenter (Hopper)
H100 PCIe 350W 80GB HBM3 Datacenter $260/year
H100 SXM 700W 80GB HBM3 Datacenter $521/year
H200 SXM 700W 141GB HBM3e Datacenter $521/year
NVIDIA Datacenter (Blackwell)
B100 700W 192GB HBM3e Datacenter $521/year
B200 1,000W 192GB HBM3e Datacenter $744/year
B300 1,100W 288GB HBM3e Datacenter $819/year
GB200 NVL72 2,700W/tray Combined Superchip $2,010/year
AMD Instinct
MI300X 750W 192GB HBM3 Datacenter $558/year
MI325X 750W 256GB HBM3e Datacenter $558/year
MI355X 900W 288GB HBM3e Datacenter $670/year
Intel
Gaudi 3 900W 128GB HBM2e Datacenter $670/year

*Annual cost assumes 85% utilization, PUE 1.5, $0.10/kWh, 24/7 operation. Actual costs vary by configuration.

Cooling Requirements

Heat Generation Basics

All electrical power consumed by GPUs converts to heat that must be removed. Use the formula: BTU/hr = Watts × 3.412

Example: 8× H100 SXM Server
  • GPU Power: 8 × 700W = 5,600W
  • Heat Output: 5,600 × 3.412 = 19,107 BTU/hr
  • Cooling Tons: 19,107 ÷ 12,000 = 1.6 tons

Cooling Technology Options

Air Cooling

Traditional CRAC/CRAH units. Suitable up to ~25kW/rack. Lower efficiency at high densities.

Up to 25 kW/rack

Rear Door Heat Exchangers

Water-cooled doors on rack rear. Extends air cooling to higher densities.

25-40 kW/rack

Direct Liquid Cooling (DLC)

Cold plates directly on GPUs. Required for highest density deployments.

40-100+ kW/rack

Immersion Cooling

Servers submerged in dielectric fluid. Maximum heat removal capability.

100+ kW/rack

Rack Density Planning

Calculate cooling needs by rack configuration:

Rack Density and Cooling Requirements
Configuration Power/Rack BTU/hr Cooling Method
2× 8-GPU H100 servers ~12 kW ~41,000 Air (Hot/Cold Aisle)
4× 8-GPU H100 servers ~24 kW ~82,000 Air + RDHx
4× 8-GPU B200 servers ~40 kW ~136,000 Direct Liquid Cooling
DGX GB200 NVL72 ~120 kW ~409,000 Liquid Cooling Required

Electrical Infrastructure Requirements

Voltage Requirements

AI GPU servers typically require higher voltage power distribution for efficiency:

  • 208V 3-phase: Most common in North American data centers
  • 240V single-phase: Common for smaller deployments
  • 400V 3-phase: European standard, increasingly used in US for efficiency

Higher voltage = lower amperage = smaller conductors = lower infrastructure cost

NEC Compliance (80% Rule)

Per National Electrical Code, continuous loads (running 3+ hours) must not exceed 80% of circuit rating. This means circuits must be sized at 125% of the load:

Circuit Size = Load Amps × 1.25
Example: A server drawing 24A continuous requires a 30A circuit minimum (24 × 1.25 = 30A)

Common Server Power Requirements

Server Electrical Requirements
Server Type Power Draw Amps @ 208V Circuit Needed
4× RTX 6000 Pro Server ~2.0 kW ~10A 15A circuit
8× A100 Server ~4.0 kW ~19A 30A circuit
8× H100 SXM Server ~6.0 kW ~29A 40A circuit
8× B200 Server ~10.0 kW ~48A 60A circuit
DGX H100 ~10.2 kW ~49A 60A circuit

PDU and Rack Power

Plan Power Distribution Units (PDUs) based on total rack load:

  • Basic PDU: Power distribution only, no monitoring
  • Metered PDU: Displays total power consumption
  • Monitored PDU: Per-outlet monitoring, remote access
  • Switched PDU: Remote power cycling capability

For AI deployments, use monitored or switched PDUs for visibility and management. Plan for N+1 redundancy on critical workloads.

Regional Electricity Rates

Electricity costs vary significantly by region, impacting total operating costs

Regional Electricity Rates Comparison
Region Avg. Commercial Rate Annual Cost (8× H100)* Notes
Texas (ERCOT) $0.065/kWh $26,900 Lowest US rates, renewable options
Virginia (Dominion) $0.075/kWh $31,000 Major data center hub
Ohio/Indiana $0.080/kWh $33,100 Growing DC market
US Average $0.100/kWh $41,400 Baseline reference
California (PG&E) $0.180/kWh $74,500 High rates, renewable mandates
New York (ConEd) $0.200/kWh $82,700 Highest US metro rates
Iceland $0.045/kWh $18,600 100% renewable, cool climate
Norway $0.050/kWh $20,700 Hydropower, free cooling
Germany $0.250/kWh $103,400 Highest European rates

*Annual cost for 8× H100 SXM at 85% utilization, PUE 1.5, 24/7 operation. Rates are approximate commercial/industrial averages and vary by contract.

Power Efficiency Best Practices

1

Optimize PUE

Reducing PUE from 1.6 to 1.3 saves 18.75% on total power costs. Invest in efficient cooling, use free cooling when possible, and maintain hot/cold aisle containment.

2

Right-Size Deployments

Match GPU selection to workload requirements. Using RTX 6000 Pro for inference instead of H100 saves 50% power per GPU while often meeting latency requirements.

3

Use Power Capping

NVIDIA GPUs support power capping via nvidia-smi. Reducing H100 from 700W to 500W often provides 80% of performance at 70% of power consumption.

4

Schedule Workloads

Run batch training during off-peak hours when electricity rates are lower. Many utilities offer time-of-use rates with 30-50% savings at night.

5

Monitor Continuously

Use DCIM tools and GPU monitoring (nvidia-smi, AMD SMI) to track actual power consumption. Identify idle GPUs and optimize utilization.

6

Consider Location

Electricity costs vary 4× between regions. For large deployments, colocating in Texas or Nordic countries can save millions over 5 years.

Frequently Asked Questions

How much power does an NVIDIA H100 GPU consume?

The NVIDIA H100 SXM consumes 700W TDP (Thermal Design Power), while the H100 PCIe variant uses 350W. Actual power consumption varies based on workload, typically ranging from 40-100% of TDP during AI inference and training tasks. An 8-GPU H100 SXM server can draw approximately 5.6kW from GPUs alone, plus additional power for CPUs, memory, storage, and networking.

What is PUE and why does it matter for AI infrastructure?

PUE (Power Usage Effectiveness) measures total facility power divided by IT equipment power. A PUE of 1.5 means for every 1kW of GPU power, 0.5kW goes to cooling and infrastructure. Modern efficient data centers achieve PUE of 1.1-1.3, while average facilities run 1.5-1.8. Lower PUE directly reduces operating costs - improving from 1.6 to 1.3 saves nearly 19% on electricity.

How do I calculate cooling requirements for GPUs?

Convert GPU power to BTU/hr by multiplying watts by 3.412. For cooling tons, divide BTU/hr by 12,000. An 8-GPU H100 server (5.6kW) produces approximately 19,100 BTU/hr, requiring about 1.6 tons of cooling capacity. For high-density deployments (40kW+ per rack), direct liquid cooling or immersion cooling becomes necessary as air cooling reaches its practical limits around 25-30kW per rack.

What electrical infrastructure do I need for AI GPUs?

AI GPU servers typically require 208V or 240V three-phase power. Per NEC code, circuits should be sized at 125% of continuous load (80% rule). An 8-GPU H100 server drawing 5.6kW needs approximately 27 amps at 208V or 23 amps at 240V, requiring a 35A or 30A circuit respectively. Plan for redundant power feeds (A+B) and monitored PDUs for production deployments.

How much does it cost to run AI GPUs annually?

Annual electricity costs depend on GPU power, utilization, PUE, and local rates. A single H100 SXM at 85% utilization with PUE 1.5 and $0.10/kWh costs approximately $7,800/year. An 8-GPU server would cost around $62,000/year in electricity alone. Over 5 years, electricity typically represents 30-40% of total cost of ownership for AI infrastructure.

What percentage of AI infrastructure TCO is electricity?

Electricity typically represents 30-40% of total cost of ownership (TCO) over a 5-year period for AI infrastructure. For high-density GPU deployments running 24/7, this can exceed 50% of TCO, making power efficiency and electricity rates critical factors in deployment decisions. Location selection (Texas vs. California can differ 3× in power costs) significantly impacts long-term economics.

Need Help Planning Your AI Infrastructure?

From hardware procurement to facility planning, SLYD provides end-to-end support for your AI deployment

GPU Hardware

Source H100, H200, B200, and MI300X systems from tier-1 OEMs including Dell, HPE, Supermicro, and Lenovo.

Explore Hardware →

Equipment Financing

Flexible financing options to spread GPU infrastructure costs over 24-60 months with competitive rates.

Learn About Financing →

Cloud Compute

On-demand GPU compute from our marketplace of verified providers. No hardware procurement delays.

Browse Compute →

Consulting Services

Expert guidance on infrastructure design, facility planning, and deployment optimization.

Contact Us →
Reconnecting to the server...

Please wait while we restore your connection

An unhandled error has occurred. Reload 🗙