Data Centers Are Booming, But It's Just the Beginning VIEW IN BROWSER Amid all the buzz about artificial intelligence (like today’s news that OpenAI hit a $500 billion valuation), one critical piece of the puzzle stands above the rest: data centers. Data centers are physical buildings that contain IT infrastructure where data is stored, managed and analyzed. They power AI, making them the cornerstone of the entire AI Revolution. Currently, there are more than 5,000 data centers in the U.S., but the reality is for the AI Revolution to continue, we need a whole lot more of them. So, in today’s Market 360, I’ll explain why and also reveal the company that benefits the most from the buildout of data centers. Plus, I’ll share where the real money will be made. (Hint: It comes after the data center boom and could trigger a $20 trillion economic wave.) Why We Need Data Centers Believe it or not, data centers have actually been around since the 1940s, with facilities housing massive computing machines. In fact, here’s a picture of NASA’s mission control computer room with dual IBM 7090s, taken in 1962.  As computers became smaller and more efficient, these rooms were known as “servers.” But with the onset of cloud computing in the early 2000s, businesses no longer needed to physically house their IT resources on site. As a result, the first server rooms evolved to today’s data centers. However, most of them are not equipped with nearly enough processing power to deal with the speed and sheer volume of data needed to train the latest AI models. They need to be at least six times more powerful… which requires a whole different set of chips, servers and other equipment. For example, AI data centers require specialized cooling units due to the heat generated by the servers; however, it’s not possible to retrofit these pre-AI data centers to handle the increased load. That’s where the “hyperscale data” center comes in. According to International Business Machines Corp. (IBM), a hyperscale data center is a massive data center that provides extreme scalability capabilities and is engineered for large-scale workloads with an optimized network infrastructure, streamlined network connectivity and minimized latency. | Recommended Link | | | | A brand new technology is lining up to be more disruptive than the internet or even today’s most advanced artificial intelligence. Those who prepare now could see massive stock gains. Those who don’t could find themselves on the wrong side of history. Click here for 3 steps to take today. |  | | Why Scale Matters Now To keep up with the accelerating demand for AI, machine learning, and cloud computing, massive data centers spanning hundreds of thousands of square feet to hundreds of acres are being constructed worldwide. Demand is skyrocketing. According to McKinsey & Company, global demand for data center capacity is anticipated to grow between 19% and 22% annually between 2023 and 2030.  To put that in dollar terms, they project that companies will invest almost $7 trillion in data center infrastructure globally. That kind of growth doesn’t happen in “normal” industries. It’s a once-in-a-generation buildout – and the companies supplying the backbone of this revolution stand to make fortunes. The Company Cashing In on the Data Center Boom No other company is cashing in more from the data center boom than NVIDIA Corporation (NVDA). Over just the past two weeks, the company has announced two major investments in the data center buildout: one with Intel Corporation (INTC) and one with OpenAI. First, NVIDIA plans to invest $5 billion into Intel to develop PC and data center chips. Together, the two companies will custom design and manufacture “multiple generations” of custom chips, as well as focus on “seamlessly connecting NVIDIA and Intel architectures” using interconnect technology, NVLink. NVLink allows several graphics processing units to connect to each other or to central processing units for fast and efficient communication between chips in data centers. The NVLink interconnect also helps to boost scaling to handle intense AI and high-performance computing workloads. NVIDIA CEO Jensen Huang commented, “This historic collaboration tightly couples NVIDIA’s AI and accelerated computing stack with Intel’s CPUs and the vast x86 ecosystem – a fusion of two world-class platforms.” Then, on September 22, NVIDIA revealed its plans to invest up to $100 billion in OpenAI. The company expects the partnership will enable OpenAI to deploy “at least 10 gigawatts” of compute capacity from NVIDIA’s AI systems to train and run OpenAI’s next-generation AI models. Huang noted that 10 gigawatts equates to between four million and five million graphics processing units (GPUs). To put this into perspective, NVIDIA will ship between four million and five million GPUs this year. So, as Huang stated, “This is a giant project.” Beyond NVIDIA’s direct investments, we’re now seeing the broader tech sector making similar moves to secure GPU access… Last month, Microsoft Corporation (MSFT) committed $17.4 billion to Nebius – a fast-growing AI “neocloud” provider. These neocloud providers lease out AI computing power, providing access to high-performance GPUs optimized specifically for AI workloads. Nebius isn’t the only GPU-as-a-Service (GPUaaS) company Microsoft has inked deals with, either. All told, it’s committed more than $33 billion to neocloud providers. While NVIDIA isn’t directly a party to this deal, Microsoft will gain access to more than 100,000 of NVIDIA’s latest GB300 chips. So, simply put, whether it’s Intel, OpenAI or Microsoft, tech companies are racing to secure NVIDIA’s technology – and the scale is unlike anything we’ve seen before. What Comes Next However, the next phase of AI isn’t about data centers; it’s “Physical AI” – robotics, automation and infrastructure powered by new AI breakthroughs – that will scale across every industry. Analysts estimate this shift could ignite a $20 trillion economic wave, reshaping logistics, retail, agriculture, manufacturing and more. My InvestorPlace colleagues Eric Fry, Luke Lango and I call it AI Day Zero. We believe the profit potential here is massive, and we don’t want investors to miss out. So, we decided to team up and create a brand-new portfolio to take advantage of it: The Day Zero Portfolio – seven carefully chosen stocks we believe are best positioned to power and profit from this once-in-a-generation transformation. If you want to be on the right side of this historic moment, I encourage you to watch our urgent AI Day Zero broadcast here. Sincerely, | .png)
.png)
Tidak ada komentar:
Posting Komentar