AMD Breaks $1 Billion in Datacenter GPU Sales in Q2 2024

AMD has achieved a major milestone by breaking $1 billion in datacenter GPU sales in Q2 2024, driven by the success of its Instinct MI300 series GPUs.

Introduction

AMD has once again demonstrated its prowess in the competitive GPU market by surpassing $1 billion in datacenter GPU sales during the second quarter of 2024.

This remarkable achievement is primarily driven by the soaring demand for AMD’s Instinct MI300 series GPUs, part of its “Antares” line of compute engines.

As AMD continues to scale its operations and enhance its product offerings, the company projects that it will surpass $4.5 billion in sales of these devices by the end of the year.

NUS Unveils CHARM3D: A Revolutionary Technique for Fabricating 3D Circuits – techovedas

AMD Achieves Record Datacenter GPU Sales

In an impressive feat, AMD has surpassed $1 billion in datacenter GPU sales in Q2 2024. The company’s Instinct MI300 series GPUs, part of its “Antares” line, have driven this significant revenue milestone. AMD now projects over $4.5 billion in sales for these GPUs by the end of the year.

Rapid Growth of Antares GPUs

Since the launch of AMD’s Epyc CPUs in 2017, the company has seen a steady revenue climb. However, the growth of Antares GPUs has been extraordinary.

Despite some initial concerns about demand earlier this year, AMD’s supply still struggles to meet the high market demand.

High Demand and Competitive Pricing

The persistent high demand for GPUs keeps their prices elevated. This trend is comparable to Nvidia’s pricing for its “Hopper” H100 and H200 GPU accelerators.

With Nvidia’s upcoming “Blackwell” B100 and B200 GPUs expected to command even higher prices, AMD’s competitive positioning remains robust.

AI Boom Sparks 75% Surge in DRAM and NAND Revenues to $150 Billion in 2024 – techovedas

Supply Chain Challenges and Progress

Lisa Su, AMD’s CEO, highlighted the company’s achievements during a call with Wall Street analysts. “On the supply side, we made great progress in the second quarter,” Su said.

“We ramped up supply significantly, exceeding $1 billion in the quarter. I think the team has executed really well.”

Despite improvements, Su acknowledged ongoing tightness in the supply chain expected to persist through 2025.

Partnerships and Capacity Expansion

AMD has built strong partnerships across its supply chain to address these challenges. The company is increasing capacity and capability to keep pace with growing demand.

Su emphasized the importance of these partnerships in meeting both supply and demand opportunities, accelerating customer adoption as the year progresses.

Projected Sales and Market Dynamics

AMD anticipates approximately $5 billion in GPU sales for 2024, primarily driven by the MI300X compute engines.

The MI300A units also contribute to this projection. The supply chain improvements from partners like Taiwan Semiconductor Manufacturing Co. (TSMC) and HBM memory suppliers SK Hynix and Samsung play a crucial role in meeting these targets.

Nvidia’s Diverse Portfolio

While AMD focuses on its high-performance GPUs, Nvidia maintains a more diverse portfolio.

Nvidia’s datacenter GPUs, including the H100 and H200, support a range of applications from HPC simulation to AI training and inference.

Nvidia’s business benefits from this balanced revenue stream, unlike AMD’s more focused approach.

AI Inference Traction

AMD’s GPUs are gaining significant traction in AI inference workloads. The MI300X’s high performance and large HBM memory make it an attractive option compared to Nvidia’s H100, which requires more GPUs due to smaller memory sizes.

Su noted that AI inference has become a primary driver for AMD’s GPU revenues, especially for GenAI workloads requiring multiple GPUs for timely responses.

Balancing AI and HPC Applications

While AMD’s GPUs are effective for AI inference, they are also suitable for AI training and HPC applications.

In recent quarters, about half of AMD’s GPU revenues came from HPC, with the current quarter seeing a shift towards AI, particularly inference.

This balance is expected to reflect the broader market trends over time.

Macronix Unveils World’s First 3D NOR Flash : Faster, Stronger, Smarter – techovedas

Profitability Prospects

The profitability of AMD’s GPU business remains a key focus. Su and CFO Jean Hu expressed confidence that the GPU business will eventually become more profitable than AMD’s average, though it has yet to reach that point.

Q2 Financial Performance

In Q2 2024, AMD reported $5.84 billion in sales, an 8.9% increase from the previous year.

Net income surged by nearly tenfold to $265 million, or 4.5% of revenues. Despite investments in new products, AMD maintained a cash reserve of $5.34 billion.

Datacenter Group Performance

The Datacenter Group, encompassing CPUs, GPUs, DPUs, and FPGAs, achieved $2.83 billion in sales, up 114.5% year-over-year. Operating income soared by a factor of 5.1 to $743 million, with CPUs contributing more significantly to this growth.

This group now comprises nearly half of AMD’s revenues, reflecting the strong performance of Epyc CPU sales and the rising demand for Antares GPUs.

Samsung Unveils GDDR7: 40% Faster and 20% More Energy Efficient Than GDDR6 – techovedas

Challenges and Future Outlook

Despite impressive growth, AMD faces challenges in maintaining its supply chain and meeting market demand. However, with continued partnerships and capacity expansions, the company is well-positioned to capitalize on the growing AI and HPC markets.

Looking ahead, AMD aims to balance its revenue streams and enhance profitability, driven by its innovative GPU and CPU offerings.

Follow us on Twitter: https://x.com/TechoVedas

Conclusion

AMD’s record-breaking Q2 datacenter GPU sales mark a significant milestone for the company.

As it continues to innovate and expand its market presence, AMD is poised for sustained growth and increased market share in the highly competitive GPU industry.

Kumar Priyadarshi
Kumar Priyadarshi

Kumar Priyadarshi is a prominent figure in the world of technology and semiconductors. With a deep passion for innovation and a keen understanding of the intricacies of the semiconductor industry, Kumar has established himself as a thought leader and expert in the field. He is the founder of Techovedas, India’s first semiconductor and AI tech media company, where he shares insights, analysis, and trends related to the semiconductor and AI industries.

Kumar Joined IISER Pune after qualifying IIT-JEE in 2012. In his 5th year, he travelled to Singapore for his master’s thesis which yielded a Research Paper in ACS Nano. Kumar Joined Global Foundries as a process Engineer in Singapore working at 40 nm Process node. He couldn’t find joy working in the fab and moved to India. Working as a scientist at IIT Bombay as Senior Scientist, Kumar Led the team which built India’s 1st Memory Chip with Semiconductor Lab (SCL)

Articles: 2237