Recommended Free Tools
Some links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
Reliable data centers now have to solve three connected engineering problems: where electricity comes from, how heat is removed, and how power quality is maintained when the grid or load changes. The sound approach is not to select a universally “greenest” technology, but to match grid and renewable procurement, cooling architecture, and UPS capacity to the site climate, water constraints, IT load profile, reliability tier, schedule, and owner priorities. The 2015 examples that inspired this topic remain useful as case studies, while current U.S. demand is documented in the Department of Energy’s 2025 update.
Why power, heat and UPS decisions belong in one design conversation
Electricity consumption makes these decisions a national infrastructure issue as well as a facilities issue. The Lawrence Berkeley National Laboratory’s 2025 update for the U.S. Department of Energy estimates that data centers used 192 TWh in 2024, equivalent to 4.7% of total U.S. electricity consumption. Its reference case models 649 TWh, or 11.8% of forecast U.S. electricity use, in 2030. The report gives a compounded uncertainty range of 521–843 TWh for 2030, so those figures are scenarios rather than measured outcomes. See the DOE/LBNL 2025 update for the assumptions and modeling.
As an Amazon Associate I earn from qualifying purchases.
Higher demand makes efficiency valuable, but efficiency is not the only objective. An owner may give priority to operating cost, a corporate emissions mandate, available utility capacity, customer requirements, construction speed, resilience, or water limitations. The best design therefore balances energy, reliability, environmental impact and financial risk instead of optimizing one metric in isolation.
The original Building Design + Construction report was published on December 1, 2015. Its projects and quotations are historical examples, not statements about current contracts or equipment. The report’s central lesson still applies: architecture, electrical engineering, mechanical systems and operations must be designed together. Dennis Julian of Integrated Design Group put it plainly: “Integrating architecture and engineering design is key.” (Building Design + Construction, Dec. 1, 2015.)
Popular on Amazon
- 1500VA/1000W PFC Sinewave Uninterruptible Power Supply (UPS): Uses sine wave output to provide battery backup power for Active PFC & conventional power supplies; Safeguards computers, workstations, network devices, and telecom equipment
- 12 NEMA 5-15R OUTLETS: 6 battery backup & surge protected outlets, 6 surge protected outlets; INPUT: NEMA 5-15P right angle, 45 degree offset plug with 5 foot power cord; 2 USB charge ports (1 Type-A, 1 Type-C) quickly charge phones and tablets
- MULTIFUNCTION, COLOR LCD PANEL: Displays immediate, detailed information on battery and power conditions; Color display alerts users to potential issues before they can affect critical equipment and cause downtime; Screen tilts up to 22 degrees
- AUTOMATIC VOLTAGE REGULATION (AVR): Corrects minor power fluctuations without switching to battery power; UL SAFETY CERTIFIED: Product has been tested in a UL certified lab and listed with UL as meeting or exceeding safety standards
- 3-YEAR WARRANTY – INCLUDING THE BATTERY; $500,000 Connected Equipment Guarantee; FREE PowerPanel Management Software (Download)
Alternative power strategies for continuous data-center loads
Power planning should distinguish the physical source of electrons from contractual and accounting arrangements. A wind-power purchase agreement can support a company’s renewable-energy goals without physically feeding the contracted electrons directly to a particular data hall. On-site generation and storage can affect the facility’s instantaneous load, but their output depends on equipment, weather, fuel, controls and interconnection limits.
| Strategy | What it provides | Questions to resolve before selection |
|---|---|---|
| Utility grid with upgraded service | Continuous supply through the local transmission and distribution system, subject to utility capacity and outage exposure. | Can the utility deliver the required megawatts on the project schedule? What substation, feeder, protection and interconnection work is required? |
| Renewable power contract | Long-term procurement of renewable generation or associated attributes, often supporting emissions goals and price planning. | Is the contract physical or financial? What delivery point, term, volume, settlement, curtailment and renewable-attribute treatment apply? |
| On-site photovoltaic generation | Local generation that can offset daytime purchases and provide a visible sustainability feature. | Is there enough unshaded roof or land? How will output vary by season, weather and building load? Does interconnection allow export? |
| Battery or other storage | Ability to shift some grid purchases, manage peaks, support ride-through or participate in demand-response programs. | What duration, cycling, fire protection, degradation, controls and replacement plan are needed? Is the storage intended for energy shifting or reliability? |
| Demand flexibility and efficiency | Lower or shifted demand through IT, cooling, controls and operating practices, reducing the amount of generation and grid capacity required. | Which loads can be curtailed without violating service-level agreements, thermal limits or redundancy requirements? |
The 2015 report described two corporate examples. HP contracted to purchase 112 MW from a SunEdison wind farm for five Texas data centers under a 12-year agreement. Novartis’ Fort Worth facility had a 100 kW photovoltaic array that also shaded parking and seating areas. Both examples illustrate different approaches—off-site procurement and on-site generation—and both are historical reporting from 2015.
Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallThe same article associated Facebook’s Fort Worth data center with the Shannon wind farm in Clay County, Texas, where 119 turbines were preparing to come online in 2015. That description should not be read as a current statement about the project’s operation or contract status.
A practical power-planning sequence
- Establish the load envelope. Document present IT demand, expected rack density, growth phases, cooling parasitics, peak demand, minimum operating load and the load that must remain online during maintenance or an outage.
- Map the utility path. Obtain written information on available capacity, energization dates, voltage, substation work, interconnection studies, tariffs, demand charges and outage history.
- Define reliability functions. Separate continuous generation, short-duration ride-through, standby generation, storage and load-shedding roles. Do not assume that a renewable contract replaces physical backup.
- Screen renewable options. Compare contract term, volume, price structure, renewable attributes, deliverability and counterparty risk with the site’s emissions objectives.
- Model on-site generation and storage together. Test hourly weather, battery state of charge, generator dispatch, peak shaving, black-start requirements, maintenance outages and extreme-weather cases.
- Check permitting and construction sequence. Interconnection equipment, transformers, switchgear, fuel systems, fire protection and control networks can determine the critical path.
- Validate operations. Write procedures for islanding, transfer, curtailment, fuel replenishment, battery replacement, manual override and restoration, then test them with the operations team.
DOE’s broader framing includes grid infrastructure, clean generation, storage, demand flexibility and efficiency. Treat these as a portfolio: a project may need several of them to satisfy both reliability and emissions requirements.
How data centers control heat
Cooling removes heat produced by IT equipment while maintaining temperature, humidity, airflow and filtration within the equipment manufacturer’s allowable range. The correct comparison depends on climate, water availability, contamination risk, heat-reuse goals, IT hardware, maintenance capability and the required operating envelope.
Also popular on Amazon
- 425VA/260W Standby Uninterruptible Power Supply (UPS): Uses simulated sine wave output to provide battery backup power and to safeguard home office, home entertainment including computers, gaming consoles, and broadband routers
- 8 NEMA 5-15R OUTLETS: Four battery backup & surge protected outlets; Four surge protected outlets; INPUT: NEMA 5-15P right angle, 45 degree offset plug with five foot power cord
- ADDITIONAL FEATURES: LED status light indicates Power-On and Wiring Fault, transformer-spaced outlets
- GREENPOWER UPS HIGH EFFICIENCY DESIGN: Reduces power consumption by utilizing a compact charger and power inverter to create an ultra-efficient backup power system for home and office use
- 3-YEAR WARRANTY – INCLUDING THE BATTERY; 75K USD Connected Equipment Guarantee; UL SAFETY CERTIFIED: Product has been tested in a UL certified lab and listed with UL as meeting or exceeding safety standards
Free-cooling and economizer systems use favorable outdoor conditions to reduce compressor or chiller work. A direct air economizer brings outdoor air into the computer room. An indirect economizer transfers heat through a separating surface so the IT space remains isolated from outdoor contaminants. Evaporative systems use water evaporation to improve heat rejection, while liquid-cooling arrangements move heat closer to high-density components. Waterless heat exchangers can reduce water dependence but may require more surface area, fan energy or favorable temperature conditions.
Julian warned in the 2015 report: “Direct outside air can be used when conditions are correct, but can also bring in contaminated air and increase operational risks.” He contrasted that with indirect systems: “Indirect cooling systems allow the computer room to stay self-contained and eliminate many of these risks.” Those comments describe the risk trade-off, not a universal ranking.
| Cooling approach | Potential advantages | Constraints and checks |
|---|---|---|
| Direct air economizer | Can reduce mechanical cooling when outdoor conditions are suitable. | Requires filtration, humidity control, contaminant monitoring and a response plan for smoke, dust, chemicals or poor outdoor-air quality. |
| Indirect economizer | Keeps the IT environment separated from outside air while using favorable outdoor temperatures. | Heat-exchanger effectiveness, fan power, fouling, controls and maintenance access affect performance. |
| Evaporative cooling | Can improve heat rejection and reduce compressor energy in appropriate climates. | Water availability, treatment, blowdown, plume, scaling, biological control and drought restrictions must be evaluated. |
| Liquid cooling | Moves heat efficiently from high-density equipment and can support higher rack loads. | IT compatibility, leak detection, fluid management, service procedures, distribution hardware and heat-reuse interfaces add design requirements. |
| Waterless heat exchange | Options such as Kyoto wheels, air-to-air plate heat exchangers and heat pipes can reduce dependence on cooling water. | Climate suitability, pressure drop, fan energy, wheel or plate maintenance, frost protection and available plant space require analysis. |
Water scarcity can make a waterless or low-water approach more attractive even when an evaporative system has a favorable energy profile. Conversely, a dry system may increase fan or mechanical energy in a particular climate. A credible comparison reports both energy and water consequences rather than labeling one technology “most efficient.”
Heat-control design steps
- Characterize the IT heat source. Record rack density, equipment inlet-temperature limits, airflow direction, liquid-cooling requirements and expected future loads.
- Analyze the local climate. Use design-day and annual weather data, smoke and pollution exposure, humidity, freeze conditions and extreme-heat events.
- Set water objectives. Establish allowed potable and non-potable water use, water quality, drought restrictions, treatment requirements and discharge limits.
- Choose the air boundary. Decide whether direct outdoor air is acceptable, whether indirect separation is required, and how filtration and contamination alarms will operate.
- Design air management. Control supply and return paths, containment, bypass airflow, fan speed, pressure relationships and sensor placement before sizing additional cooling capacity.
- Evaluate heat recovery. Identify nearby processes or buildings that can use recovered heat, then verify temperature level, seasonal demand, distance, controls and ownership of the interface.
- Commission under realistic conditions. Test economizer changeover, high-temperature operation, smoke or contamination response, water-system alarms, pump failure, control-network loss and maintenance modes.
DOE’s 2024 Best Practices Guide for Energy-Efficient Data Center Design treats environmental conditions, air management, cooling, electrical systems, heat recovery and metrics as parts of one design process. That integrated approach is more useful than selecting cooling equipment from a single efficiency number.
UPS systems: match topology and capacity to the real load
A UPS protects critical IT equipment from voltage disturbances and interruptions while the facility transfers to another source or rides through a short event. Capacity, redundancy and efficiency must be based on the actual load profile, not only the nameplate rating of installed IT equipment.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Also popular on Amazon
- 1500VA / 900W RELIABLE BACKUP POWER: The highest VA capacity available for home use; delivers short-term battery power to keep essential devices powered during blackouts, surges, and unexpected power interruptions
- TEN PROTECTED OUTLETS: Power your entire setup with 5 battery backup outlets for essential devices, and 5 surge-only outlets for peripherals. Plus built-in coaxial and Ethernet surge protection for added peace of mind
- AUTOMATIC VOLTAGE REGULATION (AVR): Corrects low voltage brownouts (88V+) and surges (+/-13%) without draining battery. Boosts or trims to stable 120V. Extends runtime for blackouts; Active PFC compatible for gaming PCs
- REPLACEABLE BATTERY & ENERGY STAR UPS: User-replaceable battery (APCRBC124, sold separately) for zero-downtime swaps. ENERGY STAR certified for 92%+ efficiency, cutting energy costs vs standard UPS units
- LCD DISPLAY PANEL: Features an intuitive LCD screen that displays real-time status information including battery charge level, estimated runtime, load capacity, and input voltage for easy monitoring of your power protection system
The 2015 article reported that Page was using rotary-type UPS systems in place of lead-acid battery systems. It also noted expert commentary that newer UPS products had improved efficiency at full load and at the lower loads common in redundant systems. This is historical commentary, not a current product comparison or a specification for any particular manufacturer.
| Decision factor | What to document |
|---|---|
| Load profile | Normal, peak, minimum, growth-phase and maintenance loads; power factor; harmonic behavior; and the effect of workload migration. |
| Redundancy architecture | Required availability, concurrent-maintenance capability, fault domains, bypass arrangements and the number of modules that can be unavailable. |
| Efficiency at operating load | Efficiency at the loads the system will actually carry, including lightly loaded redundant modules—not only the headline full-load value. |
| Energy storage and ride-through | Required ride-through duration, transfer sequence, recharge time, temperature limits, fire protection and replacement strategy. |
| Maintenance and lifecycle | Access, bypass safety, service intervals, spare parts, training, monitoring, disposal and total cost over the planned life. |
| Integration | Compatibility with generators, switchgear, building controls, emergency power-off logic, protective coordination and commissioning tests. |
UPS selection procedure
- Measure rather than guess. Use interval data or a defensible engineering model for current and projected loads, including cooling and power-distribution losses.
- Define the failure sequence. Describe what happens for utility loss, generator start failure, a UPS module fault, bypass operation, battery degradation and a maintenance event.
- Compare architectures at partial load. Calculate losses and usable capacity for normal and redundant states so an oversized system does not operate inefficiently for years.
- Coordinate protection. Verify breaker settings, short-circuit ratings, selective coordination, grounding, arc-flash requirements and generator-UPS interactions.
- Plan storage maintenance. Set inspection, testing, replacement, environmental and end-of-life procedures for batteries or other storage media.
- Commission failure modes. Demonstrate transfers, bypass, alarms, controls, communications and recovery with the actual operating team.
The available sources do not establish current comparative specifications for commercial UPS products. Obtain those values from current engineering and manufacturer documentation, and require vendors to state test conditions, load points, redundancy assumptions and warranty boundaries.
Use metrics without mistaking one number for the whole design
Power Usage Effectiveness (PUE) is the ratio of total facility energy to energy delivered to IT equipment. A lower PUE indicates fewer overhead losses under the measured boundary, but it does not by itself describe water consumption, outage performance, emissions, embodied carbon or the quality of the IT service. The 2015 report said Page was giving greater attention to reducing PUE.
| Metric or measure | Why it matters | Qualification to record |
|---|---|---|
| PUE | Shows facility overhead relative to IT energy. | State the measurement boundary, period, weather and IT-load conditions. |
| IT load and peak demand | Determines utility capacity, UPS sizing, generator requirements and demand charges. | Separate measured values from forecast growth and design margins. |
| Water use | Reveals the operational impact of evaporative and water-treatment systems. | Report source water, reclaimed water, seasonal variation and discharge assumptions. |
| Carbon or emissions intensity | Connects procurement and generation choices to corporate or regulatory goals. | Identify location-based versus contract-based accounting and the period used. |
| Availability and maintainability | Tests whether the designed redundancy can be operated and serviced as intended. | Use documented failure tests, maintenance windows and incident records. |
| Heat-recovery output | Shows whether recovered heat is actually useful rather than merely theoretically available. | Measure temperature, flow, seasonal demand and delivered useful energy. |
DOE’s 2024 guide includes efficiency metrics and benchmarking, but the evidence does not support one universally sufficient metric for comparing energy, water, reliability and emissions. Establish a dashboard that keeps those dimensions visible together.
Integrating the systems during design and construction
Many failures originate at interfaces: a renewable source that cannot meet the required physical reliability, an economizer that conflicts with contamination controls, or a UPS that is efficient only at a load the facility rarely reaches. Assign one integrated owner for the load model and require electrical, mechanical, architectural, controls and operations teams to use the same assumptions.
Also popular on Amazon
- 700VA/370W Slim Profile Standby Uninterruptible Power Supply (UPS): Uses simulated sine wave output to provide battery backup power and to safeguard home office, home entertainment including computers, gaming consoles, and broadband routers
- 8 NEMA 5-15R OUTLETS: Five battery backup & surge protected outlets, Three surge protected outlets; two outlets are widely spaced to accommodate larger plugs; INPUT: NEMA 5-15P right angle, 45 degree offset plug with five foot power cord
- 2 USB CHARGING PORTS: Share 2.4 amps to charge and power tablets, smartphones, MP3 players, and other mobile devices; LED STATUS LIGHTS: indicates Power-On and Wiring Fault
- GREENPOWER UPS HIGH EFFICIENCY DESIGN: Reduces power consumption by utilizing a compact charger and power inverter to create an ultra-efficient backup power system for home and office use
- 3-YEAR WARRANTY – INCLUDING THE BATTERY; $100,000 Connected Equipment Guarantee; FREE PowerPanel Management Software (Download); UL SAFETY CERTIFIED: Product has been tested in a UL certified lab and listed with UL as meeting or exceeding safety standards
- Freeze assumptions in a basis-of-design document: IT growth, rack density, climate files, water limits, reliability objective, utility milestones and operating modes.
- Run joint scenarios: utility outage during extreme heat, generator maintenance during a high-load period, smoke event during economizer operation, and a cooling-control network failure.
- Coordinate physical space: reserve routes and access for transformers, UPS modules, heat exchangers, pumps, filters, storage, maintenance bypasses and future expansion.
- Specify controls as a system: define sensor accuracy, alarm priorities, fail-safe states, cybersecurity responsibilities, manual overrides and trend retention.
- Commission progressively: factory-test equipment, verify installation, test subsystem interactions, then conduct integrated systems testing with realistic load banks and failure sequences.
- Turn design intent into operations: provide sequences of operation, single-line diagrams, water-treatment procedures, contamination response, battery plans and operator training.
Construction schedules also influence the choice. Utility upgrades, long-lead switchgear, transformers, generators, cooling towers, heat exchangers and UPS equipment can dominate the critical path. A phased design may allow an initial data hall to open while later capacity, renewable procurement or heat-reuse connections are completed, but each phase must preserve the promised redundancy.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Historical case studies and what they actually show
The 2015 report described Novartis’ 37,000-square-foot, $53 million Fort Worth data center on the Alcon Laboratories campus, opened September 29, 2015, with the 100 kW PV array noted above. It also described the Texas wind projects and HP agreement as examples of corporate approaches available at that time. These figures establish the scope of those reported projects; they do not provide a current cost benchmark, performance result or operating-status claim.
The report also quoted Andy Baxter, Principal and MEP Engineering Director at Page: “They just have a different set of business priorities.” That observation explains why two owners facing similar loads may select different combinations of contracts, generation, cooling and UPS systems. Mike Silla of Skanska’s Mission Critical Center of Excellence said, “Companies that were building data centers on a large scale are now building what they need, when they need it,” describing the 2015 market context rather than a current forecast.
Common failure modes and corrective actions
Counting renewable procurement as firm backup
Symptom: The sustainability plan shows renewable megawatt-hours, but the resilience plan has no physical source during a grid outage. Correction: Document the contract structure separately from on-site generation, storage, UPS ride-through and standby generation.
Sizing cooling from average conditions
Symptom: Economizer or heat-rejection capacity works in typical weather but fails during design-day heat, smoke or humidity. Correction: Test annual and extreme conditions, changeover logic, filtration, water restrictions and degraded equipment states.
Also popular on Amazon
- 1500VA/900W Intelligent LCD Uninterruptible Power Supply (UPS): Uses simulated sine wave technology to provide battery backup power to safeguard workstations, networking devices, and home entertainment equipment
- 12 NEMA 5-15R OUTLETS: Six battery backup & surge protected outlets; six surge protected outlets; INPUT: NEMA 5-15P plug with 6-foot power cord; USB charge ports (1 Type-A, 1 Type-C) quickly charge mobile phones and tablets
- MULTIFUNCTION, COLOR LCD PANEL: Displays immediate, detailed information on battery and power conditions; Color display alerts users to potential issues before they can affect critical equipment and cause downtime
- AUTOMATIC VOLTAGE REGULATION (AVR): Corrects minor power fluctuations without switching to battery power; UL SAFETY CERTIFIED: Product has been tested in a UL certified lab and listed with UL as meeting or exceeding safety standards
- 3-YEAR WARRANTY – INCLUDING THE BATTERY; 500,000 Connected Equipment Guarantee; FREE PowerPanel Personal Software (Download)
Ignoring low-load UPS efficiency
Symptom: A redundant UPS operates far below its rating and produces avoidable losses. Correction: Compare efficiency and operating modes at the actual normal, minimum and maintenance loads, then align module count and expansion stages with the load plan.
Optimizing PUE while increasing water or risk
Symptom: A lower energy overhead is achieved with unacceptable water demand, contamination exposure or maintenance complexity. Correction: Report energy, water, reliability and emissions together and obtain owner approval for the trade-off.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Leaving controls and commissioning until the end
Symptom: Equipment meets individual specifications but fails during transitions or combined faults. Correction: write sequences early, include controls in procurement, and schedule integrated systems testing before occupancy.
A decision checklist for owners and design teams
- What IT load must be served today, at peak, at minimum and after each expansion phase?
- Which loads must remain online for each failure and maintenance scenario?
- What utility capacity, interconnection work and energization date are documented?
- Is renewable power being procured for emissions accounting, price management, physical delivery, or several purposes?
- What are the site’s climate, air-quality, water-availability and drought constraints?
- Does direct outdoor air introduce contamination risks that the owner is prepared to manage?
- Can the chosen cooling system support current and future rack densities and any liquid-cooling interfaces?
- What are the UPS efficiency, redundancy and maintenance requirements at the actual operating load?
- How will storage, generators, UPS, cooling and controls behave during a combined extreme event?
- Which metrics will be measured after handover, by whom, and under what boundary and reporting period?
There is no single best combination of alternative power, heat control and UPS technology. A defensible design states its assumptions, quantifies the trade-offs, tests failure modes and keeps the facility adaptable as IT demand and grid conditions change. That is the practical interpretation of the historical case studies and the current DOE guidance.
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.
