By Claire Simmers, Ph.D.
Originally Printed in DE Sierra Magazine 2025 (Truncated Version)
Well, another year has passe,d and I’m back with you. Last year I wrote about the Energy Enigma and this year, there is another somewhat related topic I’d like to write about – the Data Center Enigma. I don’t have a degree in computer science or information technology, but my experience is probably very similar to yours – there is a lot of data and many information technology devices in my life. I spend countless hours on my devices for work, organizations like the Sierra Club, and for personal uses. I want– ok, I need–my data to keep flowing, but how does this happen? Data centers play a huge part in making this happen for all of us. Data centers use extraordinary amounts of resources, particularly energy and water and create noise pollution. Could I live without my access to the digital world? Could any person or business exist in the present and future of our professional and personal lives without data centers? Let’s see if we can untangle the Data Center Enigma.
BACKGROUND OVERVIEW
According to the latest estimates from Statista, throughout 2024, approximately 402.74 million terabytes of data are generated daily (equal to approximately 402.74 quintillion bytes – 4.0274 × 10²⁰), encompassing newly created, captured, copied, and consumed information. Data is measured using a hierarchical system of units, starting from bytes as the basic building block. A single byte consists of 8 bits, each representing a binary value (0 or 1). As data volume increases, it is expressed in larger units for ease of understanding: kilobytes (KB), megabytes (MB), gigabytes (GB), terabytes (TB), petabytes (PB), exabytes (EB), and zettabytes (ZB). Each unit represents a thousandfold increase from the previous one (e.g., 1 TB = 1,000 GB). For extremely large data volumes- yottabytes (YB), the next unit after zettabytes, are occasionally referenced. The volume of data in the world is growing at an unprecedented rate. By 2025, global data is projected to reach 181 zettabytes, with significant contributions from AI-driven content, social media and user-generated content, Internet of Things (IoT) devices, enterprise and transactional data, scientific research, Big Data, cloud computing, digital storage, eCommerce, streaming, and digital transactions.
The average person's monthly data usage varies significantly by activity, but generally falls between 5-20 GB for mobile data and can exceed 1 TB for heavy home internet users. Activities like streaming HD video and 4K video consume the most data, while activities like web browsing, music streaming, and email use far less. The average large business enterprises have 347.56TB of data, seven times as much data as the average small business with 47.81TB. Similarly, organizations of all sizes expect the amount of data they will use to increase considerably in a relatively short timeframe.
This is BIG DATA. Big data is data so large that traditional data processing methods just can’t handle it. Billions of people in the world now have internet access. Every action they take online generates new data. Businesses across every industry want to use that data to collect customer information, track inventory, manage human resources, and more. In fact, we create data at such an alarming rate that we’ve had to invent new words like zettabyte to measure it. All those trillions of gigabytes of data need to be stored, analyzed, and kept secure. Enter the Big Data Industry. Big Data couldn’t exist without data centers to store and transmit data from all over the world. A data center refers to a building or group of buildings that house servers, hardware, networking equipment, and other computing technologies. These data centers offer vital cloud, managed, and colocation data services. There are 5,376 data centers in the US alone. Most US data centers are in or around the leading tier 1 metro markets. The most popular US data center locations are Northern Virginia and Northern California, with key markets such as those in Virginia and Silicon Valley. While data centers are a critical linchpin in Big Data, they create BIG challenges.
DATA CENTER CHALLENGES
High Energy Consumption: data centers are massive consumers of electricity. Large data centers can consume over 100 megawatts (MW), enough to power 80,000 US households. US data centers used approximately 176 terawatt-hours (TWh) of electricity in 2023, which was 4.4% of the nation's total electricity use. The energy demand from data centers is expected to grow rapidly, potentially increasing to 6.7% to 12% of US electricity use by 2028.
High Water Consumption: data centers can use millions of gallons of water daily for cooling, straining local resources. Large data centers can consume up to 5 million gallons per day, equivalent to the water use of a town populated by 10,000 to 50,000 people. According to scientists at the University of California, Riverside, each 100-word AI prompt is estimated to use roughly one bottle of water (or 519 milliliters). This may not sound like much, but billions of AI users worldwide enter prompts into systems like ChatGPT every minute. Large language models require many energy-intensive calculations, necessitating more liquid cooling systems.
Greenhouse Gas Emissions: The energy consumption at data centers contributes to significant CO2 and greenhouse gas emissions. About 56% of the electricity used to power data centers nationwide comes from fossil fuels. Data centers’ projected electricity demand in 2030 is set to increase to up to 130 GW (or 1,050 TWh), which would represent close to 12% of total US annual demand. Building new fossil-fuel plants to fulfill this demand will increase carbon emissions and further contribute to climate change.
Noise Pollution: data centers produce noise that can be heard up to two miles from the facility, with the worst noise being up to 3,000 feet from the facility. The noise can have significant health implications for nearby residents and wildlife, and the low-frequency noise from the cooling fans cannot be blocked using traditional sound barriers.
Electronic and Toxic Waste (E-waste): The rapid obsolescence of hardware leads to a growing amount of electronic waste, which is increasing at an alarming rate and is a major sustainability challenge. Data centers generate significant amounts of e-waste, including servers, storage, and networking equipment, which contain hazardous materials like lead, mercury, and cadmium. This waste poses environmental risks by polluting landfills and leaching toxins into the environment.
I hope I have established that data centers fulfill a critical need that is not going away, and, in fact, the need is growing, AND that they present tremendous challenges – the Data Center Enigma. What is to be done?
SOLVING THE DATA CENTER ENIGMA
We all have a role to play – data center developers, legislators, scientists, and the general populace.
DATA CENTER DEVELOPERS
High Energy Consumption and Greenhouse Gas Emissions: data center developers can build them in places with an abundance of renewable energy, they can generate renewable power on-site, and they can make the facilities more energy efficient, thereby reducing the need for power from the grid.
High Water Consumption: developers can instead use innovative water management techniques to reduce water consumption, including closed-loop cooling systems, immersion cooling, and using non-potable water sources (e.g., recycled wastewater and captured water).
Closed-loop cooling systems enable the reuse of both recycled wastewater and freshwater, allowing water supplies to be used multiple times. A cooling tower can use external air to cool the heated water, allowing it to return to its original temperature. These systems can reduce freshwater use by up to 70%.
Free cooling is a method whereby outside cold air is drawn into the data center to cool the equipment. Data centers must be located in cooler climates for this strategy to be effective.
Immersion cooling in data centers involves bathing servers, chips, and other components in a specialized dielectric (or non-conductive) fluid. Hardware is submerged in specially designed tanks filled with the coolant. The non-conductive liquid absorbs the heat from the chips and transfers it to a heat exchanger, where it is cooled down before flowing back into the tank. Immersion cooling is a novel process that entails higher upfront costs than conventional direct liquid cooling but provides significant energy savings and space-optimization benefits for data center developers. Since the technology uses synthetic fluids, it requires significantly less water than other approaches.
Powering data centers with renewable energy sources, like solar or wind, requires significantly less water consumption than obtaining energy from fossil-fuel power plants.
Noise Pollution: data center developers reduce non-low-frequency noise pollution by using passive strategies like soundproofing enclosures for generators and HVAC units, installing acoustical louvers, using sound-absorbing materials, and employing passive noise cancellation with strategic site planning. Developers can also strategically position equipment to isolate it from neighboring properties and create natural sound barriers with landscaping. Active strategies involve using quieter fans and implementing immersion cooling systems to reduce or eliminate the need for noisy fans, thereby reducing or eliminating low-frequency noises that do not respond to traditional soundproofing techniques.
Electronic and Toxic Waste (E-waste): To address this, data center developers should implement comprehensive lifecycle strategies focusing on reducing waste, extending equipment life, and partnering with certified recyclers for proper and secure destruction and disposal.
LEGISLATORS
Legislators can control data center problems by introducing bills to increase transparency, reform tax incentives, and manage energy and water usage. Other strategies include passing legislation to protect ratepayers from infrastructure costs, ensuring grid reliability through demand response programs, and enacting stricter zoning and siting regulations to mitigate environmental impacts.
Transparency & Reporting: mandate disclosure by requiring data center developers to publicly report their water and power usage to localities before construction, allowing for better planning and impact assessment. Legislators can also require developers to disclose if they are working on similar projects in the area to avoid double-counting issues.
Energy & Grid Management: use cost allocation schemes by creating data center-specific utility tariffs or rate structures that ensure these facilities pay for the energy and grid infrastructure they require, preventing costs from being shifted to other ratepayers. Implement Demand Response Programs that compensate large data centers for reducing their electricity consumption during peak grid stress to improve reliability and prevent blackouts. Require investments in the modernization of the electrical grid to handle increased demand from data centers and ensure a stable energy supply for all users.
Tax Incentives & Economic Oversight: review or revise existing tax exemptions for data centers to cap the amount of tax revenue lost or to study their fiscal impact more closely. Promote Direct Energy Sourcing by enabling power suppliers to contract directly with data centers to meet their energy needs, potentially avoiding the need for massive grid build-outs paid for by general customers.
Siting & Environmental Regulation: implement zoning reform by overhauling data center siting and zoning laws to allow for more centralized, state-level oversight, or to standardize zoning rules to treat data centers like other industrial uses. Increase environmental protection by implementing regulations to mitigate noise pollution, air pollution, and water usage impacts associated with data centers. Improve site selection by encouraging the use of brownfield sites and other previously developed areas for new data centers to minimize their environmental footprint.
Fund Educational & Research Initiatives: Legislators can support educational and research efforts by enacting state and federal laws that mandate studies on data center impacts, establish working groups to develop solutions, and create dedicated funds for renewable energy and efficiency research linked to data center growth. They can also direct universities and state agencies to research sustainable data center practices, such as improving energy and water efficiency, exploring renewable energy integration, and developing specialized energy tariffs for data centers.
SCIENTISTS
Scientists can reduce data center pollution by working on energy efficiency, innovating sustainable cooling and heat recovery, and developing policies and monitoring tools for oversight.
1. Improve energy efficiency by developing more efficient hardware. Research into low-power processors, energy-efficient power supplies, and solid-state drives can substantially reduce equipment energy consumption. Scientists can use AI and machine learning to analyze energy usage, predict demand, and dynamically allocate workloads to the most efficient servers. This allows facilities to power down unused equipment during periods of low demand. Research can focus on developing more efficient software architectures and virtualization technologies. These can consolidate multiple virtual machines onto fewer physical servers, maximizing resource use and reducing the total number of machines required.
2. Innovate sustainable cooling and heat recovery by advancing solutions for heat management and developing technologies that reuse waste heat. Liquid cooling is more efficient than traditional air cooling, especially for high-density computing loads like AI. Scientists can work on new technologies such as those listed previously. With nearly 97% of a data center's electrical energy turning into heat, scientists can create new ways to capture and repurpose this waste. Applications include: using recovered heat for district heating systems for nearby residential and commercial buildings, creating energy through Organic Rankine Cycle (ORC) turbines, which turn heat into electricity, supporting agricultural needs, such as heating greenhouses or water for fish farms.
3. Develop policy, monitoring, and circular economy tools by providing data and models to inform policy decisions and create better resource management systems. Scientists can advocate for informed policy, quantify the environmental impact of data centers, and provide data to help policymakers craft effective regulations. This includes policies that mandate efficiency standards, require the use of renewable energy, or regulate water consumption. They can advance Data Center Infrastructure Management (DCIM) and AI tools to monitor energy and water consumption in real-time, helping facilities operate more efficiently. Metrics like Power Usage Effectiveness (PUE) and Water Usage Effectiveness (WUE) can be continually optimized. Research is needed to improve equipment lifecycle management, including better recycling programs for e-waste. This reduces the need for new materials and minimizes waste going to landfills.
4. Transition to sustainable energy sources by pushing for the development and integration of clean energy sources. Advance on-site energy solutions like more efficient solar panels and battery storage to reduce reliance on the grid. Model and optimize data center energy consumption to align with the availability of renewable energy on the power grid. This is known as "carbon-aware computing," and it can shift workloads to times and locations where clean energy is most abundant.
THE GENERAL POPULACE (aka, you and me)
The general populace can reduce harm from data centers by supporting policies that mandate transparency, energy and water efficiency, and renewable energy use, while also participating in local planning processes to advocate for community-friendly locations and impacts. By demanding greater accountability and promoting sustainable growth, communities can mitigate negative effects on their environment and public health.
Advocate for Policy and Transparency: support legislation that requires data centers to disclose their energy and water consumption, allowing for better public oversight and informed action. Advocate for policies at the state and local level that mandate energy audits, require data centers to use renewable energy, and align their growth with climate goals. Encourage local governments to conduct health and environmental impact assessments before approving data center construction and to ensure public input is a part of the decision-making process.
Influence Community Planning: participate in community meetings and forums to express concerns and advocate for measures that protect the local community from potential harms, such as noise and emissions. Advocate for policies that require data centers to be located a sufficient distance from sensitive areas like residential neighborhoods and schools.
Foster Broader Awareness: share information about the energy and water demands of data centers and advocate for responsible recycling of data center e-waste to promote sustainability. Encourage investment in research and development for more efficient cooling technologies, which can significantly reduce a data center's water and energy footprint.
BIG Data is vital, creating BIG Challenges requiring BIG Efforts from all of us.
References:
Bartley, K. 2025 (May 28). Big data statistics: How much data is there in the world? Retrieved from: https://tinyurl.com/34szh2mc.
Yañez-Barnuevo, M. 2025 (April 15). Data Center Energy Needs Could Upend Power Grids and Threaten the Climate. Retrieved from: https://www.eesi.org/articles/view/data-center-energy-needs-are-upending-power-grids-and-threatening-the-climate
Yañez-Barnuevo, M. 2025 (June 25). Data Centers and Water Consumption. Retrieved from: https://www.eesi.org/articles/view/data-centers-and-water-consumption