15 MASSIVE Data Centers Under Construction

[00:00:00] Speaker 1: Around the world, these new high-performance data centers are under construction. They're going to be designed to deliver unprecedented speed and processing power for artificial intelligence applications. So in today's video, we're going to take a look at 15 of the largest data centers currently being built. Let's start with number 15, the Meta AI Data Center in Louisiana. In a major step for both technology and regional development, Meta, Facebook's parent company, is building its largest AI data center yet in northeastern Louisiana. It's located on a 2,200-acre site between Rayville and Delhi, and this $10 billion project marks a huge leap forward in AI infrastructure, while also delivering a much-needed economic boost to that region. At 4 million square feet, this facility will be one of the largest hyperscale AI data centers in the world. Now, these hyperscale centers are designed to handle massive amounts of data, powering advanced tech like AI training, deep learning, and real-time data analysis. These facilities are going to be crucial for supporting the growing demands of AI-driven innovations, from machine learning to natural language processing. Now, Meta is also focusing on sustainability here, with plans to integrate about 1,500 megawatts of renewable energy into Louisiana's grid in the future. However, the project currently relies on a partnership with Intergy to build three natural gas power plants, which will supply energy to the facility for the next 15 years. This has sparked some controversy, with environmental groups raising concerns about the increased use of fossil fuels. Well, Meta's journey from college dorm room to startup in 2000 floor to a global tech giant has been remarkable. Today, the company is a leader in AI and virtual reality, and this Louisiana facility highlights the immense computing power that's needed to support that digital ecosystem. For Richland Parish, one of Louisiana's most economically struggling areas, this project is a lifeline. With a population of just 20,000 and a heavy reliance on agriculture, this region has long-faced high poverty rates and limited job opportunities. This data center is expected to create 500 permanent high-paying jobs and another 1,000 indirect jobs through related economic activity. And during construction, the project is going to employ 5,000 workers, providing a significant short-term boost to the local economy. And Meta is also investing $200 million in local infrastructure, including roads and water systems, to ensure that the project benefits the community as well as the company. Number 14. SDC Austin Campus in Texas The SDC Austin Campus in Round Rock, Texas, is going to be a groundbreaking leap forward in the data center infrastructure, supercomputing, and sustainable technology sector. Developed by Savvy Data Centers, one of the largest privately-owned data center operators in the U.S., this facility is built to meet the growing demands of hyperscale and high-performance computing. These are HPC applications. It's a major addition to the global network of advanced data centers. It's located about 4 miles from downtown Round Rock, and it's about 18 miles from Austin. This 40-acre campus sits on the site of a former Sears call center, now transformed into a cutting-edge data center hub. When fully completed, this campus will feature two buildings totaling around 430,000 square feet, with the capacity to support up to 84 megawatts of critical IT power. This first building, completed in mid-2023, is already operational and equipped to handle some of the world's most energy-intensive applications, including liquid-cooled supercomputers. The second building, spanning 105,000 square feet, is going to further expand the capacity for hyperscale and enterprise clients, solidifying this campus as a key tech hub in that region. What sets the SDC Austin campus apart is its ability to support high-density cooling, up to 200 kilowatts per cabinet. This makes it ideal for next-gen supercomputers like Horizon, the largest academic supercomputer in the National Science Foundation's portfolio. Horizon will power open scientific research across multiple fields, showing the campus' role in driving academic and technological innovation. The project was born out of Texas' growing demand for large-scale, high-density computing facilities. In February of 2022, the Round Rock City Council approved the project, offering tax incentives in exchange for Sabi's commitment to invest at least $185 million in construction, $500 million in business equipment, and create 20 jobs over five years. 13. University of Alabama High Performance Computing Center The University of Alabama is poised to become a leader in advanced computing with the construction of its new high-performance computing and data center. It was approved by the university. This project is set to be complete by late 2026, on the eastern edge of the UA campus in Tuscaloosa. It's designed to house the most powerful computing system in Alabama. The facility is going to aim to enhance research capabilities across the state and beyond. The facility was also going to provide $25 million worth of cutting-edge computing equipment alongside additional infrastructure to support cloud storage solutions. Now, this center is going to serve multiple purposes, primarily supporting research across all campuses of the university. Its computational power will drive advances in key research areas like water resource management, where large-scale systems can be simulated and analyzed, and quantum computing, which will explore new frontiers in secure communications and computational theories. Other key areas include health and medicine, accelerating drug discovery and modeling complex biological systems, transportation, optimizing logistics and developing smart mobility solutions, material science, innovating new materials with enhanced properties, and national security, supporting classified research requiring secure and controlled data processes. The total cost of this project, including the construction and equipment, is estimated to exceed $50 million. Now, while this represents a significant investment, the economic returns are expected to far outweigh the initial expenditure. By enabling this research and fostering partnerships with industry leaders, the HPC Center is projected to stimulate economic development within Alabama, creating new opportunities in technology-driven sectors. So, DataBank's ambitious new project, this is a three-story-tall, purpose-built data center in Plano, Texas. It's slated to set a new benchmark for scale and capacity. This facility is going to form part of DataBank's broader expansion strategy. Now, that's driven by record demand for its services and significant investment from major stakeholders. The new data center is being constructed on a six-acre area adjacent to an existing DataBank facility at Dominion Parkway in Texas. This marks DataBank's first three-story, ground-up data center, designed to produce 40 megawatts of power. That's a critical metric reflecting its ability to handle high-density workloads. This vertical construction approach was chosen due to space constraints on the available land, but beyond Plano, DataBank is undertaking 15 additional data center projects in locations like Northern Virginia, Minnesota, Salt Lake City, and Atlanta. DataBank is headquartered in Dallas, Texas, and it operates over 65 data centers across the U.S. The company's expansion is rooted in a strategy to address the growing supply constraints in the U.S. data center market. With the existing facilities in the Dallas-Fort Worth region operating at full capacity, this project is a direct response to the urgent need for additional space and power. Australian Super has invested about $2 billion as part of a $3 billion equity raise to support DataBank's development. This marks a significant move by the Australian Fund to capitalize on the growing demand for digital infrastructure. Now, each client typically requires 3 to 9 megawatts of power, which underscores this facility's 40 megawatt capacity as a critical asset. This center is going to house thousands of high-performance servers optimized for AI workloads, cloud services, and enterprise applications. Here's hoping it gives me 60 FPS in Cyberpunk. Number 11. The MGH-PCC The Massachusetts Green High-Performance Computing Center stands as a hub for high-performance computing in Holyoke, Massachusetts. This facility that's connected to the city's municipal fiber grid and powered by renewable energy from the Holyoke Canal system represents a collaborative effort of six prominent universities – Boston University, Harvard, MIT, Northeastern, the University of Massachusetts System, and Yale. Established in 2012, the center has continuously expanded its capacity, and its recent announcement in October of 2024 signals an ambitious leap into quantum computing. This facility has a computing capacity that's going to support hundreds of thousands of processing cores across its clusters. By 2016, one UMass cluster alone featured over 14,376 cores from Intel and AMD, and over 1.1 petabytes of storage on a cutting-edge FDR InfiniBand network. Now, this facility has a current size of about 90,000 square feet and room for expansion, with over four acres of undeveloped land available. In October, they announced a $16 million expansion aimed at developing a quantum computing complex. Now, this project's co-funded by Massachusetts, contributing about $5 million, and Boston-based QERA computing, which invested about $11 million. QERA plans to deploy its neural atom quantum computer at the center. Now, this is going to mark a milestone in quantum research accessibility for academic institutions. The entire thing was conceived in the late 2000s as a response to the increasing need for centralized high-performance computing facilities in New England. Its founding members recognized the potential of pooling resources to address these large-scale challenges. Holyoak was chosen for its cost-efficient, renewable energy infrastructure and proximity to member institutions. This quantum computing is an entirely different paradigm from classical computing. QERA's neutral atom technology relies on rubidium atoms, controlled and manipulated using lasers to serve as qubits. These systems are going to perform calculations leveraging superposition and entanglement, enabling them to solve problems in combinatorial optimization and condensed matter physics that are infeasible for classical systems. Founded in 2018 and rooted in research from MIT and Harvard, the company's current flagship system features 256 qubits, and it operates on Amazon's cloud platform, Bracket. QERA specializes in analog quantum computing, but it aims to develop hybrid systems, combining analog and digital. Its research focuses on areas like the Reitberg atoms, optimization problems, and condensed matter physics. The Tract Data Center Park, located in Buckeye, Arizona, is poised to become one of the largest in the United States. This project spans over 2,000 acres in Maricopa County, within the Phoenix metropolitan area. Buckeye, the largest city in Arizona by area, is a strategic choice for this development. Tract is a relatively new player in the data center industry. It's officially launched in 2023. This company is led by Grant von Royden, an experienced leader in digital infrastructure. It specializes in developing large-scale, master-planned data center parks across the U.S., and they've got a portfolio that includes over 23,000 acres in states like Nevada, Utah, and Virginia. The data center is expected to encompass up to 20 million square feet of data center space, distributed across 40 individual facilities. The infrastructure is going to aim to support up to 1.8 gigawatts of power, making it one of the most advanced data center parks globally. The site has been zoned for industrial use, which is going to enable rapid development of data center facilities, and it's designed to accommodate up to that 1.8 gigawatts. The park relies on partnerships with local utilities to meet those energy demands. Additionally, the park's going to be integrated into high-speed fiber optic networks, ensuring low-latency connectivity for this hyperscale computing operation. The Phoenix area is already top-tier data center market, offering favorable conditions such as proximity to major energy infrastructure, a stable climate, and access to high-speed networks. Buckeye's specific advantages include this location near the Palo Verde generating station and other renewable energy initiatives, which will help support the park's immense power requirements. 9. The Jupiter Exascale Supercomputer So, Jupiter, joint undertaking pioneer for innovative and transformative exascale research, marks Europe's ambitious entry into the exascale computing era, a transformative milestone in the world of high-performance computing. As the first European exascale computer, Jupiter is set to redefine computational capability. The project is hosted at the ULIC Supercomputing Center in Germany, a fitting location for a system designed to tackle some of humanity's most complex challenges. The initiative was championed by Euro-HPC Joint Undertaking, and that shows their commitment to leadership in computational innovation. Established in 2018, they laid the groundwork for Jupiter through a comprehensive funding strategy. With a budget of about 273 million euros, the project secured the resources necessary for its goals, and by January 2023, a call for tender initiated the construction phase, and installation of the campus began in early 2024. Now, at its core, Jupiter is a marvel. Its dual-module design integrates two distinct computational modules, each optimized for specific tasks. The booster module, equipped with nearly 24,000 NVIDIA GPUs, no wonder I can't get a 50-90, is engineered for highly scalable, intensive problems such as artificial intelligence training and scientific simulations. To support the architecture, it employs those NVIDIA, Mellanox, and FiniBan NDR networking tech, and this high-performance network ensures seamless communication between the modules. Its performance exceeds one exaflop. Now, that's a scale of power that's almost unimaginable, equivalent to every person on Earth performing one calculation per second, continuously for over four years, all condensed into a single second. This extraordinary capability is made possible through developed hardware and software. It's an example of the continent's growing expertise in high-performance computing. However, achieving these feats is not going to be without its challenges. Constructing a supercomputer of this magnitude demands intricate planning. Cooling the system, for example, poses a considerable challenge given the heat generated by its components, but to address it, Jupiter incorporates some advanced cooling tech designed to minimize any environmental impact while also maintaining its efficiency. That's pretty good. 8. ALPS Supercomputer The ALPS Supercomputer, inaugurated in September of 2024, is a high-performance facility located in Switzerland under the Swiss National Supercomputing Center. Funded by the Swiss Confederation through the F domain, ALPS is poised to redefine computational capacity. It's going to offer 270 petaflops of processing power. That's equivalent to 270 quadrillion operations per second, ranking sixth on the global top 500 list of supercomputers. CSCS was founded in 1991 with the mission of supporting Switzerland's scientific community through computational services. This project was proposed to address the escalating need for extreme performance, particularly for AI applications in medicine, climate research, and physics. The ALPS system is housed in a purpose-built data center. Now, this facility spans three floors, with the lowest dedicated to power distribution, emergency batteries, and water cooling. That cooling system pumps about 460 liters per second from a depth of about 45 meters through a two-kilometer-long pipeline. The computing system occupies the top floor, with an installation area of about 21,000 square feet. ALPS is powered by those NVIDIA Grace Hopper GH200 circuits, which combine about 72 Grace CPUs and 132 Hopper GPUs into a single chip. Yeah, that'll run Quake. Each processing node contains four GH200 chips and a highly interconnected structure, offering a combined 1.3 million CPU and GPU cores across over 2,600 nodes. Running the ALPS supercomputer requires immense resources. It consumes 10 megawatts of electricity under foot load. The annual electricity costs are going to be estimated at $15 million, with the total project cost reaching an approximate $100 million. Well, the center's engineers are working closely with researchers to develop software tailored to specific applications, which will enable more efficient utilization of its capability. But ALPS differentiates itself with its cloud-native architecture, which is going to support the creation of software-defined clusters that are tailored to specific user needs. This flexible approach ensures that researchers can maintain confidentiality while maintaining computational efficiency. Moving on to number 7, Amazon Data Centers in Ohio Amazon Web Services is currently investing $10 billion to expand its data center in Ohio, signaling an ambitious step towards reshaping the landscape of cloud computing. These centers are poised to be some of the top supercomputer facilities under construction in the U.S., placing Ohio at the forefront of data processing power. These data centers, with the projected completion date in late 2026, will encompass a 590-acre campus in Jeffersonville, Ohio, located in Fayette County. Now, these centers are more than just warehouses for computer equipment. They're going to serve as the backbone for a new wave of cloud computing services, driving innovations in AI and ML, which is machine learning. Each facility is going to house thousands of servers and provide enormous computational capacity, which is critical for everything from running cloud-based applications to processing data for AI algorithms. AWS's existing data center campuses in central Ohio and places like New Albany, Hilliard, and Dublin have already established the region as a key player in the data sector. And the company's new projects will only deepen that footprint, adding hundreds of megawatts in capacity, including the recent 400-megawatt capacity boost in 2024 alone. The scale and impact of these data centers go beyond just adding more storage space. They're going to be part of a critical infrastructure system that supports both large corporations and smaller businesses around the world. AWS data centers are known for their ability to quickly scale up thanks to their modular design, and when additional capacity is needed, new servers can be added with minimal disruption. This flexibility is crucial, as demand for computational power is going to continue to grow, especially as industries such as AI and machine learning require increasingly larger amounts of data. Number 6. Ducalion Supercomputer The Ducalion Supercomputer, situated in the Mino Advanced Computing Center in Portugal, represents a significant milestone in Europe's high-performance computing landscape. Now, it was officially inaugurated in September of 2023. It's Portugal's most powerful supercomputer, and one of the top installations of the Euro HPC network. It was designed to achieve a peak performance of 10 petaflops, or about 10 million billion calculations per second. It's a key player in advancing research, innovation, and scientific collaboration across Europe. The origins of the center trace back to 2019, when the European High Performance Computing Joint Undertaking announced plans to establish eight supercomputers in various European Union locations. These projects aim to bolster the EU's computing capability and its position as a leader in cutting-edge technology. By February of 2021, a contract to build this one was signed between Euro HPC, Portugal's Foundation for Science and Technology, and Fujitsu. Ducalion's architecture sets it apart from other supercomputers in the Euro HPC network. It features three main compute partitions, each tailored to specific workloads. The largest partition utilizes 1632 Fujitsu Prime PC FX700 nodes with ARM V8A architecture processors. This segment alone delivers a peak performance of five petaflops, supported by 16GB of high-bandwidth memory per node. The second partition is based on x86 architecture, and it comprises about 500 Atos Bull Sequana X440 nodes equipped with AMD EPYC ROM 7742 processors. The investment here is justified by the broad range of applications that this place supports. Its system is designed to advance research in fields like climate modeling, astrophysics, and artificial intelligence, even personalized medicine and fluid dynamics. This makes it a cornerstone in the European Processor Initiative, which aims to develop energy-efficient microprocessors tailored to European needs. As of mid-2024, this place has already begun making an impact, hosting about 50 research projects, with 20% of them being international collaborations under the Euro HPC framework. This supercomputer's ability to handle diverse workloads makes it a valuable resource for scientists, innovators, and industries across Europe. 5. Leonardo Supercomputer Leonardo is a pre-X-Scale supercomputer that stands as one of Europe's most significant achievements in high-performance computing. It's situated in the Bologna Technopole in Italy, and it's part of the European High-Performance Computing Joint Undertaking. It was inaugurated in November of 2022. Leonardo resides on a campus that was transformed from a mid-20th century tobacco factory into a cutting-edge research and computing hub. The supercomputer is managed by Seneca, a non-profit consortium of 70 Italian universities, national institutions, and ministries. Now, the system's composed of a Belsa-Quana HX2000 supercomputer with approximately 14,000 NVIDIA Ampere GPUs and a 200-gigabit NVIDIA Mellanox HDR InfiniBand interconnect. Now, it's divided into three distinct modules. It's optimized for varied high-performance tasks. This module delivers over 240 petaflops. It handles data-intensive workloads like artificial intelligence and machine learning, and it features 1,500 Intel Sapphire Rapid CPUs. Finally, the front-end and service module supports the two primary computing modules, with storage capabilities of 5 petabytes of high-speed storage and 100 petabytes of capacity storage, ensuring that data processing and retrieval is efficient. Combined, Leonardo can perform 250 petaflops, or 250 quadrillion, calculations per second. This places it among the top five supercomputers globally, and the second in Europe as of 2022. Half of all these computational resources are allocated to the European researchers through open calls, making it a shared resource for scientific advancement. Now, this collaborative model is intended to democratize access to high-performance computing. The hopes for it extend beyond these impressive specifications. It's expected to play a pivotal role in addressing global challenges, from improving weather forecasting accuracy to developing renewable energy technology. Leonardo represents a remarkable step forward in Europe's high-performance computing landscape. With its vast processing power, modular design, and wide-ranging applications, it's more than just a supercomputer. It's a critical enabler of progress across disciplines. 4. Equinix Hyperscale Data Centers In October of 2024, Equinix announced a groundbreaking joint venture with Singapore's Sovereign Wealth Fund, GIC, and the Canadian Pension Plan Investment Board to raise over $15 billion. The objective is to expand Equinix's footprint of hyperscale data centers in the U.S., targeting the construction of new facilities and adding over 1.5 gigawatts of capacity. Equinix, already a global leader in digital infrastructure, operates 260 data centers across 71 cities, supporting over 10,000 customers and generating $8 billion in annual revenue. The first U.S. X-scale data center will open at Equinix's Silicon Valley campus, with additional facilities planned in strategic locations. Now, these hyperscale data centers, we've talked about them before, they require significant real estate, power, and cooling infrastructure, often necessitating rural or suburban sites where land and energy are more affordable. The scale of these facilities is staggering. Their program aims to achieve 700 megawatts of capacity with an investment of over $7.5 billion when fully built out. To fund that growth, Equinix employs Joint Ventures, a model which partners like GIC and PGIM Real Estate own 80% of the equity, while Equinix retains 20%. Building and operating these data centers involves addressing engineering challenges, too. These facilities have to be equipped with advanced cooling systems to manage the heat generated by GPU-driven AI workloads. For example, Equinix recently deployed the first hyperscale liquid cooling system, essential for supporting power-dense AI infrastructure. The cost of constructing this data center can reach in the billions of dollars, depending on its size and sophistication. The need for these massive facilities stem from the exponential growth of data and the increasing reliance on AI. Training large language models and other AI systems requires unprecedented computational power, storage, and cooling. And these hyper data centers are uniquely positioned to meet those demands. These centers also support the broader ecosystem of co-location facilities and enterprise networks, ensuring some seamless data movement and connectivity. The impact of Equinix's hyperscale expansion is going to be far-reaching. By tripling its investment in these facilities, Equinix is positioning itself as a key player in the AI and cloud computing revolution. These centers will not only support the operations of tech giants, but also enable enterprises to leverage cutting-edge tech, driving innovation across industries. As AI continues to evolve, the demand for these hyperscale data centers is only going to grow, making them an essential component of this digital economy that we're now entering. 3. A Colossus Supercomputer in Memphis In June of 2024, Memphis, Tennessee, became the unexpected center of a technological revolution. Elon Musk's artificial intelligence company, XAI, announced the construction of Colossus, a supercomputer designed to be the largest and most powerful AI training platform in the world. Located at 3231 Paul R. Lowry Road on the site of the former Electrolux manufacturing plant, Colossus has quickly transformed from a bold concept to a record-breaking project. Now, the XAI company was founded in August of 2023, and by June of 2024, Memphis had been selected for its flagship project. Over just 90 days, agreements were finalized for their facility, which launched operations in September of 2024. Now, such rapid progress is unusual for a project of this magnitude, reflecting the urgency behind XAI's goals. The first phase of Colossus deployed over 100,000 NVIDIA H100 GPUs, making it a centerpiece for AI model training. By December 2024, the system had already doubled in size to 200,000 GPUs, establishing it as the world's largest supercomputer, according to NVIDIA. The Tennessee Valley Authority's agreement to provide over 100 megawatts of power underscores the immense energy needs of such an operation. Now, the supercomputer's architecture relies on cutting-edge supermicro servers with a network optimized for ultra-low latency and high bandwidth. This ensures that communication between GPUs is critical for parallel processing tasks. The facility also features advanced cooling, of course, to manage the heat generated by its densely packed servers. But unlike many supercomputers designed for academic or public sector research, Colossus is exclusively dedicated to XAI's projects. Its primary function is to train AI models, including Grok3, a chatbot integrated with X and available to premium subscribers. Additionally, Colossus is expected to train future AI systems with capabilities beyond current technology. This whole thing, Colossus, represents an investment of at least $6 billion, a figure that shows the scale of its ambition. The power consumption is already enormous, with further expansions being more than doubling its current energy requirements, while XAI has added diesel generators to the site that approach is insufficient for future phases. While the project is backed by Elon Musk, the world's richest person, whose resources and influence have allowed XAI to pursue ambitious goals, the company's exclusive focus on proprietary AI models distinguishes it from other supercomputing facilities, which are often shared among multiple users. So, as Colossus expands, it's going to face several challenges. The energy requirement of a 1 million GPU system will strain local infrastructure, and questions remain about the sustainability of the operations here. Additionally, the project's reliance on proprietary applications means that its success is tied directly to the fortunes of XAI. Industry experts speculate on whether Musk's massive AI investments will yield long-term benefits or face setbacks in a competitive and rapidly evolving market. 2. Atlas Development Data Center Campus The Atlas Development Data Center Campus is an ambitious $17 billion initiative set to become one of the world's largest data centers upon completion. It's located about 45 miles south of Atlanta, Georgia. The project, dubbed Project SAIL, shows a significant leap in global data infrastructure. This 13-building campus, with a combined footprint of nearly 5 million square feet, aims to position Georgia as a key player in the increasingly vital digital economy. The proposed site for this center spans 830 acres near the town of Nguyen. It's bordered by Wagers Mill Road and Welcom Sargent Road, and the location was chosen for its proximity to Georgia power plant Yates, a gas power station capable of producing over 700 megawatts of energy. Each of the 13 data halls will be 370,000 square feet, housing state-of-the-art computing systems to manage an enormous volume of data. With every facility expected to consume up to 72 megawatts of power, the total energy usage will exceed all of that county's households combined. The scale of this operation shows the importance of the project, but it highlights the infrastructure challenges it presents. The project was officially unveiled in late 2023 through a development of regional impact filing with Georgia's Department of Community Affairs. Since then, it's garnered significant attention due to the economic and technological implications. If completed as planned, Project SAIL is projected to generate over $1.6 billion in annual tax revenue for the county and cement the region's role as a data hub. The demand for such a massive facility stems from the growing reliance on our digital infrastructure. As industries ranging from entertainment to healthcare, they adopt cloud computing and artificial intelligence. The need for large-scale data storage and processing has skyrocketed. The $17 billion price tag shows the complexity of such an undertaking. The data center on this scale require extensive planning, not only to house the hardware, but also to ensure reliable power and cooling systems. Data centers generate immense heat, necessitating advanced cooling solutions, and furthermore, the energy demands of the facility raise questions about sustainability, especially given the environmental concerns surrounding large-scale data centers. The economic impact of Project SAIL does extend beyond its tax contributions. Data centers typically create jobs during construction and require skilled workers for maintenance and operation. However, their long-term employment opportunities are often limited compared to other industries due to the high degree of automation involved. This dichotomy between economic promise and limited job creation has sparked debates about the true benefits of these developments. Critics also point out to the strain these facilities place on local resources, particularly water and electricity, as reasons for concern. Well, despite the challenges, the significance of Project SAIL can't be overstated. Atlanta has emerged as a major data center hub in recent years, attracting companies like Google, Microsoft, and QTS to the region. This project, with its scale, positions Atlas development as a key player in that expanding market. Number 1. Stargate Project in the US The Stargate Project represents one of the most ambitious supercomputer data center initiatives ever conceived. Proposed as a $500 billion joint venture, It's spearheaded by OpenAI, SoftBank, Oracle, and the investment firm MGX. Announced in January of 2025, this venture aims to construct multiple state-of-the-art data centers across the United States, with an initial $100 billion investment already underway. This project's scale, purpose, and implications mark it as a transformative step in artificial intelligence infrastructure. Now, the Stargate Project's primary construction effort is currently focused in Abilene, Texas, where 10 data centers are being built. These facilities are expected to be some of the largest in the world, both in physical size and computing capacity. Now, the roots of the Stargate Project trace back to 2022, when OpenAI and Microsoft began discussions about building a $100 billion AI supercomputer. As AI systems like ChatGPT demonstrated their transformative potential, the need for dedicated, scalable infrastructure became apparent. The project was formally announced by President Donald Trump on January 21st, 2025, alongside SoftBank's Masayoshi-san, OpenAI's Sam Altman, and Oracle's Larry Ellison. Now, the Stargate Project isn't limited to serving a single company or purpose. Instead, it's designed as a national infrastructure project, capable of supporting various AI applications across industries. The data centers will house advanced supercomputers equipped with NVIDIA GPUs, custom-built accelerators, and next-gen AI chips developed by companies like ARM. These facilities will enable faster model training, larger-scale simulations, and real-time AI deployment. But the financial scale of Stargate is staggering. With a planned investment of up to $500 billion by 2029, it ranks among the largest infrastructure projects in U.S. history. The initial $100 billion has already been allocated to construction in Texas, and further expansion is planned in other states. The project is also expected to generate over 100,000 jobs, including roles in construction, engineering, and operations. However, such an endeavor comes with some challenges, particularly regarding energy consumption. Data centers of this magnitude are going to require vast amounts of electricity, not only for computation, but also for cooling the high-density servers. The White House has prioritized accelerating energy infrastructure development to support Stargate, with regulatory measures being streamlined to expedite the permitting process. These data centers could become hubs for innovation, enabling researchers, companies, and governments to tackle some of the most pressing challenges of our time. If realized as planned, Stargate could set a new standard for AI infrastructure, reshaping industries and enhancing the United States' position in the global technology landscape. Thanks for watching, everybody. I'm going to see if I can go grab a 5090. Thank you to our channel members.