Recommended Free Tools
Start with the expected IT load, not a facility-wide power figure. Apply an explicitly stated power usage effectiveness (PUE) assumption to estimate total facility demand, translate IT electrical load into a preliminary heat-removal requirement, and then test both estimates against the candidate site’s utility, climate, water, and expansion constraints. These are screening estimates—not a substitute for a site-specific engineering design or confirmation from the utility and local authorities.
What to estimate before comparing sites
Keep three quantities separate: IT electrical demand, total facility electrical demand, and cooling capacity. They answer different questions. IT demand is the load from servers, storage, and networking equipment. Total facility demand includes IT plus the supporting electrical loads represented in PUE. Cooling capacity is the rate of heat removal the cooling system must provide; it is not the same as the cooling system’s electrical consumption.
- Power: Usually expressed in kilowatts (kW) or megawatts (MW). Distinguish sustained demand from peak demand.
- Energy: Power consumed over time, such as kilowatt-hours (kWh) or megawatt-hours (MWh). Estimate this separately if you need operating-cost or annual-energy projections.
- Cooling capacity: A thermal rate, commonly stated in kW of cooling or refrigeration tons. It describes heat removed, not electricity used.
A preliminary estimate is only as useful as its assumptions. Record the planned equipment, utilization, deployment schedule, growth, redundancy, and whether each number represents a sustained or peak condition.
Step 1: Forecast the IT electrical load
Build an equipment and rack-load inventory
List the servers, storage, network equipment, and other IT devices expected in each deployment phase. Use the best available power figures for the actual equipment configuration and identify whether each figure is a measured or manufacturer-rated value. Sum the expected loads to produce an IT-load estimate; do not treat a rack’s power-distribution rating or a facility’s utility service rating as proof of its expected IT consumption.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errorsWhere a deployment is not fully specified, make the assumptions visible: number of racks, anticipated load per rack, expected utilization, and how quickly the equipment will be installed. Keep a sustained-load case and a peak-load case if both matter to the site decision. A utility or electrical designer may need to plan around peak demand and the timing of load increases, even when an annual-energy estimate is also available.
Model phases and growth explicitly
Calculate the load at opening and at each planned expansion milestone, rather than treating the final build-out as an immediate requirement. A site that can meet the first phase may not have enough service capacity, space, or cooling headroom for later phases. State the expected ramp-up dates and what must be delivered before each phase can operate.
Step 2: Estimate whole-facility power with PUE
Power usage effectiveness (PUE) is a dimensionless ratio: total facility energy divided by IT equipment energy. DOE FEMP describes it using annual energy in both the numerator and denominator. For an early estimate, multiplying IT power by an assumed PUE gives an approximate facility power figure when the IT and facility values refer to comparable operating conditions:
Estimated facility power = IT power × assumed PUE
For illustration only, if a hypothetical project expects 1 MW of IT load and assumes a PUE of 1.5, its estimated facility demand is 1.5 MW at that operating condition. The 1.5 PUE is an example assumption, not a recommended target or a prediction for a particular site. Replace it with a justified project assumption as the design develops.
Free tools Windows power users keep installed
One-click scans. No signup required.
PUE is not a universal design constant. Cooling architecture, climate, electrical systems, operating conditions, and the boundary used for measurement affect the result. Be clear about whether your figure is a planning assumption, a design target, or an operating measurement. Do not add a separate cooling-power estimate to the facility-power figure if the PUE calculation already accounts for facility overhead; that would count the cooling electricity twice.
Decide whether you need peak power, annual energy, or both
A single facility-power estimate does not answer every planning question. Use a peak or coincident-demand estimate to discuss service capacity and electrical infrastructure. Use annual energy for energy-use and operating-cost analysis. If the project will ramp in phases, show the expected demand by phase and when each increase is likely to occur. Tariffs, demand charges, utility service rules, and energization schedules depend on the actual site and provider; a general estimate cannot establish them.
Step 3: Translate IT load into a preliminary cooling requirement
For an early thermal balance, treat the IT equipment’s electrical consumption as heat that must be removed. Thus, 1 MW of IT electrical load represents approximately 1 MW of IT heat at the facility boundary, before accounting for other heat sources or design conditions. This is a starting point, not a complete cooling design. Add other relevant heat loads as the project is defined, and have the design team establish the applicable operating and environmental conditions.
Use consistent thermal units. One refrigeration ton is 12,000 Btu per hour, or approximately 3.517 kW of cooling. On that basis, 1 MW of IT heat is about 284 refrigeration tons before adding other loads or project-specific design allowances. The ton value describes cooling capacity, not electrical input.
Rank #3
Estimate cooling-system electrical use separately
Cooling capacity and cooling-system power are different quantities. DOE FEMP’s 2024 guide gives the following benchmarks for average cooling-system power divided by average data-center cooling load. They are reference values, not guaranteed results for a particular design:
| DOE FEMP 2024 benchmark | Cooling-system power per cooling load | Interpretation in the guide |
|---|---|---|
| Standard | 1.1 kW/ton | Standard benchmark |
| Good practice | 0.8 kW/ton | Good-practice benchmark |
| Better | 0.6 kW/ton | Better benchmark |
For the illustrative 284-ton cooling load above, applying those benchmarks would imply approximately 313 kW, 227 kW, or 171 kW of average cooling-system power, respectively. These are arithmetic illustrations using the DOE FEMP 2024 benchmark values and the rounded example load; they are not a forecast, design selection, or whole-facility power estimate. Actual system performance depends on the design and operating conditions.
Step 4: Define redundancy and spare capacity with the design team
Do not add an arbitrary percentage to the load and call it a reliability allowance. Required redundancy depends on the owner’s availability requirement, system topology, maintenance strategy, load ramp, and acceptable failure scenarios. Define what equipment must remain operational during maintenance or a component failure, and ask the electrical and cooling designers to show how that requirement affects capacity and phasing.
Keep reliability capacity distinct from expected operating load. A design may need installed spare capacity without consuming that full capacity during normal operation. The distinction matters when discussing utility service, equipment sizing, and the cost and practicality of future expansion.
Step 5: Compare cooling approaches against the site
DOE FEMP describes conventional chilled-water systems using chillers and cooling towers, air-side economizing, and direct liquid cooling approaches. None is best for every data center. Compare the options against the anticipated rack loads, local conditions, water constraints, IT environmental requirements, and the operator’s ability to maintain the system.
| Factor | What to establish for the candidate site |
|---|---|
| Climate and economizer potential | Assess local ambient conditions and realistic economizer hours. Outdoor-air cooling opportunities depend on air quality and the IT equipment’s environmental limits, not just average outdoor temperature. |
| Water availability and use | Determine whether the proposed heat-rejection method uses water, what local access and constraints apply, and whether the site can support the design through relevant seasonal conditions. Evaporative heat rejection uses water; dry heat rejection can reduce consumption but may affect energy performance and design. |
| Cooling electrical demand | Compare expected cooling-system power against cooling load using a defined metric and operating assumptions. Nominal capacity alone does not reveal how much electricity the system will use. |
| IT compatibility | Confirm that the proposed air or liquid cooling method and operating conditions are compatible with the actual equipment. DOE advises maximizing inlet temperature only while meeting IT thermal guidelines. |
| Operations and maintenance | Review equipment, controls, maintenance procedures, and operator capability. Liquid or hybrid approaches can add equipment and control sequences that the operations team must be prepared to support. |
DOE FEMP’s 2024 Best Practices Guide for Energy-Efficient Data Center Design states that no design guide can identify the most energy-efficient design for every scenario. A favorable climate, constrained water supply, high rack density, or demanding resilience requirement can shift the practical choice.
Use water and efficiency metrics carefully
Water usage effectiveness (WUE), as described by DOE, is annual site water use divided by annual IT equipment energy, expressed in liters per kWh. It can help compare water use when its measurement boundary and operating period are clear. It does not establish whether a particular site has adequate water rights, supply, or permits; those require local confirmation.
DOE FEMP’s 2019 cooling-water guidance reports that practices enabling higher chilled-water temperatures and reduced airflow were associated with a 20% reduction in chiller energy in the best-practices guide it cites. That is a source-reported opportunity, not a universal savings guarantee. The same 2019 page reports that increasing cooling-tower cycles of concentration from three to six reduces makeup-water requirements by 20% and blowdown by 50%. Those figures apply to that specific operating change, not to total data-center water use.
Quick wins for a faster PC:
Scan for outdated or missing drivers - takes under a minuteDriver Scan →Repair Windows errors before they cause bigger problemsFix Now →Step 6: Screen the site before making a commitment
Apply the estimates to each candidate site, then confirm the local facts that a generalized estimate cannot provide. DOE’s federal data-center consolidation guidance identifies climate zone, economizer hours, cooling efficiency, energy source, and expansion capacity as evaluation criteria. Its 2013 publication is a useful checklist, but current utility processes and local requirements must be verified for the actual project.
- Electrical service: Ask the utility whether the required peak demand can be served, what upgrades may be needed, and when power can realistically be energized. Confirm how the answer changes across project phases.
- Expansion: Check whether additional service, switchgear, generation, cooling equipment, and physical space can be added on the intended schedule.
- Climate and air quality: Assess seasonal conditions and air quality relevant to economizer use, alongside equipment environmental limits.
- Water: Confirm source, availability, seasonal constraints, discharge or treatment requirements, and applicable rights or permits for the intended cooling approach.
- Cooling feasibility: Test whether the preferred air, chilled-water, liquid, or hybrid architecture is practical for local conditions and the projected rack loads.
- Commercial and regulatory facts: Obtain site-specific tariffs and identify applicable permits and utility requirements. General DOE guidance cannot establish these for an unspecified location.
DOE’s tribal data-center FAQ notes that water needs vary with data-center size and cooling technology and flags water access as a planning issue. It does not provide a site-specific water entitlement or consumption estimate. Metering becomes more useful once a facility or comparable operating data exists; DOE’s metering guidance describes its role in capacity planning and energy decisions.
A concise estimate to take into site discussions
Prepare a one-page estimate with the figures and assumptions kept separate:
- IT load by phase, including expected sustained demand and peak demand.
- The assumed PUE, how it was selected, and the resulting approximate facility power for each relevant operating case.
- Preliminary heat-removal capacity, with the conversion and any additional heat loads stated.
- Cooling-system electrical-use assumptions, using an explicitly identified metric and source rather than treating cooling capacity as electrical demand.
- Redundancy and growth requirements defined with the owner and design team, not hidden inside an unexplained percentage.
- Open site questions for utility capacity and timing, climate, water, tariffs, permits, and expansion.
Use this estimate to identify infeasible sites and direct due diligence. Before selecting a site, obtain confirmation for power delivery, water access, local requirements, and the cooling architecture from the relevant providers and qualified project designers.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Quick Recap
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.
