Technology
Thursday, May 23rd, 2024 2:44 pm EDT
Key Points
- Data Center Growth: Nvidia’s data center business saw a remarkable 427% growth in the latest quarter, driven by high demand for its AI processors from major cloud providers like Amazon, Microsoft, Google, and Oracle, which contributed to 45% of Nvidia’s $22.56 billion in data center sales.
- Strong Return on Investment: Nvidia assures investors that its AI chips provide a substantial return on investment for customers, with finance chief Colette Kress highlighting that $1 spent on Nvidia hardware can generate $5 over four years. Newer products like the HDX H200 promise even higher returns, potentially generating $7 for every $1 spent.
- Future Outlook and New Technology: Nvidia plans to release its next-generation GPU, Blackwell, in the fiscal fourth quarter, aiming to maintain its momentum and address ongoing demand. Major customers for this new technology include Amazon, Google, Meta, Microsoft, OpenAI, Oracle, Tesla, and Elon Musk’s xAI. Following these developments, Nvidia’s shares surged 6%, surpassing $1,000 for the first time, and the company announced a 10-for-1 stock split after a 25-fold increase in its share price over five years.
Nvidia’s recent historic rally is being propelled by an exceptional 427% growth in its data center business in the latest quarter, fueled by surging demand for its artificial intelligence processors. This growth underscores Nvidia’s ability to not only supply advanced AI hardware but also assure investors that these investments will yield substantial returns for customers. The increasing importance of Nvidia’s AI chips is reflected in their use by major cloud providers—Amazon Web Services, Microsoft Azure, Google Cloud, and Oracle Cloud—which accounted for around 45% of Nvidia’s $22.56 billion data center sales in the April quarter.
Additionally, specialized GPU data center startups are emerging as significant customers, purchasing Nvidia’s GPUs to offer rental services for AI model training. For instance, CoreWeave charges $4.25 per hour for renting an Nvidia H100, a crucial service for training large language models like OpenAI’s GPT. Following its impressive earnings report, Nvidia’s CFO Colette Kress highlighted the strong return on investment for cloud providers using Nvidia hardware, with a $1 investment potentially generating $5 over four years. This ROI is expected to be even higher with newer Nvidia hardware, such as the HDX H200, which can generate $7 in revenue over four years per $1 spent.
CEO Jensen Huang emphasized the intense demand for Nvidia’s GPUs, with entities like OpenAI, Google, Anthropic, and thousands of AI startups eagerly seeking GPU capacity. This demand has created pressure on Nvidia to accelerate delivery and deployment of its systems. Meta’s commitment to spend billions on 350,000 Nvidia chips, despite not being a cloud provider, illustrates the vital role of AI infrastructure, termed “AI factories” by Huang, for future AI production and integration into services like advertising or chatbots within existing apps.
Nvidia also reassured analysts with an aggressive timeline for its next-generation GPU, Blackwell, set to debut in data centers by the fiscal fourth quarter. Early adopters of Blackwell include tech giants such as Amazon, Google, Meta, Microsoft, OpenAI, Oracle, Tesla, and Elon Musk’s xAI, indicating strong market anticipation and continued growth prospects.
Following these announcements, Nvidia’s shares surged by 6% in extended trading, crossing the $1,000 mark for the first time. The company also declared a 10-for-1 stock split, reflecting the monumental 25-fold increase in its share price over the past five years. This strategic move aims to make the stock more accessible to a broader range of investors, further capitalizing on its robust market performance and growth trajectory.
For the full original article on CNBC, please click here: https://www.cnbc.com/2024/05/22/nvidia-no-sign-of-ai-slowdown-after-over-400percent-jump-in-data-center-unit.html
