Thursday, August 24th 2023
Strong Cloud AI Server Demand Propels NVIDIA's FY2Q24 Data Center Business to Surpass 76% for the First Time
NVIDIA's latest financial report for FY2Q24 reveals that its data center business reached US$10.32 billion—a QoQ growth of 141% and YoY increase of 171%. The company remains optimistic about its future growth. TrendForce believes that the primary driver behind NVIDIA's robust revenue growth stems from its data center's AI server-related solutions. Key products include AI-accelerated GPUs and AI server HGX reference architecture, which serve as the foundational AI infrastructure for large data centers.
TrendForce further anticipates that NVIDIA will integrate its software and hardware resources. Utilizing a refined approach, NVIDIA will align its high-end, mid-tier, and entry-level GPU AI accelerator chips with various ODMs and OEMs, establishing a collaborative system certification model. Beyond accelerating the deployment of CSP cloud AI server infrastructures, NVIDIA is also partnering with entities like VMware on solutions including the Private AI Foundation. This strategy extends NVIDIA's reach into the edge enterprise AI server market, underpinning steady growth in its data center business for the next two years.
NVIDIA's data center business surpasses 76% market share due to strong demand for cloud AI
In recent years, NVIDIA has been actively expanding its data center business. In FY4Q22, data center revenue accounted for approximately 42.7%, trailing its gaming segment by about 2 percentage points. However, by FY1Q23, data center business surpassed gaming—accounting for over 45% of revenue. Starting in 2023, with major CSPs heavily investing in ChatBOTS and various AI services for public cloud infrastructures, NVIDIA reaped significant benefits. By FY2Q24, data center revenue share skyrocketed to over 76%.
NVIDIA targets both Cloud and Edge Data Center AI markets
TrendForce observes and forecasts a shift in NVIDIA's approach to high-end GPU products in 2H23. While the company has primarily focused on top-tier AI servers equipped with the A100 and H100, given positive market demand, NVIDIA is likely to prioritize the higher-priced H100 to effectively boost its data-center-related revenue growth.
NVIDIA is currently emphasizing the L40s as their flagship product for mid-tier GPUs, meaning several strategic implications: Firstly, the high-end H100 series is constrained by the limited production capacity of current CoWoS and HBM technologies. In contrast, the L40s primarily utilizes GDDR memory. Without the need for CoWos packaging, it can be rapidly introduced to the mid-tier AI server market, filling the gap left by the A100 PCle interface in meeting the needs of enterprise customers.
Secondly, the L40s also target enterprise customers who don't require large parameter models like ChatGPT. Instead, it focuses on more compact AI training applications in various specialized fields, with parameter counts ranging from tens of billions to under a hundred billion. They can also address edge AI inference or image analysis tasks. Additionally, in light of potential geopolitical issues that might disrupt the supply of the high-end GPU H series for Chinese customers, the L40s can serve as an alternative. As for lower-tier GPUs, NVIDIA highlights the L4 or T4 series, which are designed for real-time AI inference or image analysis in edge AI servers. These GPUs underscore affordability while maintaining a high-cost-performance ratio.
HGX and MGX AI server reference architectures are set to be NVIDIA's main weapons for AI solutions in 2H23
TrendForce notes that recently, NVIDIA has not only refined its product positioning for its core AI chip GPU but has also actively promoted its HGX and MGX solutions. Although this approach isn't new in the server industry, NVIDIA has the opportunity to solidify its leading position with this strategy. The key is NVIDIA's absolute leadership stemming from its extensive integration of its GPU and CUDA platform—establishing a comprehensive AI ecosystem. As a result, NVIDIA has considerable negotiating power with existing server supply chains. Consequently, ODMs like Inventec, Quanta, FII, Wistron, and Wiwynn, as well as brands such as Dell, Supermicro, and Gigabyte, are encouraged to follow NVIDIA's HGX or MGX reference designs. However, they must undergo NVIDIA's hardware and software certification process for these AI server reference architectures. Leveraging this, NVIDIA can bundle and offer integrated solutions like its Arm CPU Grace, NPU, and AI Cloud Foundation.
It's worth noting that for ODMs or OEMs, given that NVIDIA is expected to make significant achievements in the AI server market for CSPs from 2023 to 2024, there will likely be a boost in overall shipment volume and revenue growth of AI servers. However, with NVIDIA's strategic introduction of standardized AI server architectures like HGX or MGX, the core product architecture for AI servers among ODMs and others will become more homogenized. This will intensify the competition among them as they vie for orders from CSPs. Furthermore, it's been observed that large CSPs such as Google and AWS are leaning toward adopting in-house ASIC AI accelerator chips in the future, meaning there's a potential threat to a portion of NVIDIA's GPU market. This is likely one of the reasons NVIDIA continues to roll out GPUs with varied positioning and comprehensive solutions. They aim to further expand their AI business aggressively to Tier-2 data centers (like CoreWeave) and edge enterprise clients.
Source:
TrendForce
TrendForce further anticipates that NVIDIA will integrate its software and hardware resources. Utilizing a refined approach, NVIDIA will align its high-end, mid-tier, and entry-level GPU AI accelerator chips with various ODMs and OEMs, establishing a collaborative system certification model. Beyond accelerating the deployment of CSP cloud AI server infrastructures, NVIDIA is also partnering with entities like VMware on solutions including the Private AI Foundation. This strategy extends NVIDIA's reach into the edge enterprise AI server market, underpinning steady growth in its data center business for the next two years.
In recent years, NVIDIA has been actively expanding its data center business. In FY4Q22, data center revenue accounted for approximately 42.7%, trailing its gaming segment by about 2 percentage points. However, by FY1Q23, data center business surpassed gaming—accounting for over 45% of revenue. Starting in 2023, with major CSPs heavily investing in ChatBOTS and various AI services for public cloud infrastructures, NVIDIA reaped significant benefits. By FY2Q24, data center revenue share skyrocketed to over 76%.
NVIDIA targets both Cloud and Edge Data Center AI markets
TrendForce observes and forecasts a shift in NVIDIA's approach to high-end GPU products in 2H23. While the company has primarily focused on top-tier AI servers equipped with the A100 and H100, given positive market demand, NVIDIA is likely to prioritize the higher-priced H100 to effectively boost its data-center-related revenue growth.
NVIDIA is currently emphasizing the L40s as their flagship product for mid-tier GPUs, meaning several strategic implications: Firstly, the high-end H100 series is constrained by the limited production capacity of current CoWoS and HBM technologies. In contrast, the L40s primarily utilizes GDDR memory. Without the need for CoWos packaging, it can be rapidly introduced to the mid-tier AI server market, filling the gap left by the A100 PCle interface in meeting the needs of enterprise customers.
Secondly, the L40s also target enterprise customers who don't require large parameter models like ChatGPT. Instead, it focuses on more compact AI training applications in various specialized fields, with parameter counts ranging from tens of billions to under a hundred billion. They can also address edge AI inference or image analysis tasks. Additionally, in light of potential geopolitical issues that might disrupt the supply of the high-end GPU H series for Chinese customers, the L40s can serve as an alternative. As for lower-tier GPUs, NVIDIA highlights the L4 or T4 series, which are designed for real-time AI inference or image analysis in edge AI servers. These GPUs underscore affordability while maintaining a high-cost-performance ratio.
HGX and MGX AI server reference architectures are set to be NVIDIA's main weapons for AI solutions in 2H23
TrendForce notes that recently, NVIDIA has not only refined its product positioning for its core AI chip GPU but has also actively promoted its HGX and MGX solutions. Although this approach isn't new in the server industry, NVIDIA has the opportunity to solidify its leading position with this strategy. The key is NVIDIA's absolute leadership stemming from its extensive integration of its GPU and CUDA platform—establishing a comprehensive AI ecosystem. As a result, NVIDIA has considerable negotiating power with existing server supply chains. Consequently, ODMs like Inventec, Quanta, FII, Wistron, and Wiwynn, as well as brands such as Dell, Supermicro, and Gigabyte, are encouraged to follow NVIDIA's HGX or MGX reference designs. However, they must undergo NVIDIA's hardware and software certification process for these AI server reference architectures. Leveraging this, NVIDIA can bundle and offer integrated solutions like its Arm CPU Grace, NPU, and AI Cloud Foundation.
It's worth noting that for ODMs or OEMs, given that NVIDIA is expected to make significant achievements in the AI server market for CSPs from 2023 to 2024, there will likely be a boost in overall shipment volume and revenue growth of AI servers. However, with NVIDIA's strategic introduction of standardized AI server architectures like HGX or MGX, the core product architecture for AI servers among ODMs and others will become more homogenized. This will intensify the competition among them as they vie for orders from CSPs. Furthermore, it's been observed that large CSPs such as Google and AWS are leaning toward adopting in-house ASIC AI accelerator chips in the future, meaning there's a potential threat to a portion of NVIDIA's GPU market. This is likely one of the reasons NVIDIA continues to roll out GPUs with varied positioning and comprehensive solutions. They aim to further expand their AI business aggressively to Tier-2 data centers (like CoreWeave) and edge enterprise clients.
18 Comments on Strong Cloud AI Server Demand Propels NVIDIA's FY2Q24 Data Center Business to Surpass 76% for the First Time
There's so many super computers assembled these days with both AMD Epyc CPU's and Mi300's and what more. AMD's compute devision do better and better each year. CDNA is excellent. Its just adoptation which will take some years.
They were wrong.Supercomputing isn't just one discipline. They are typically designed, assembled, and deployed for a narrow range of applications.
Remember that CPU cores aren't suited for all workloads. Heck, the crypto craze cause high demand for GPUs because GPU cores are more suited for crypto mining than CPU cores. No one has mined bitcoin on CPUs for perhaps ten years; almost all bitcoin mining is done on custom ASICs.
Hell, 3D rasterization on CPU cores is very inefficient. That's why you have that GTX 1060 in your system. Hell, Windows doesn't even support a CPU-only graphics pipeline anymore because even the wimpiest integrated graphics processor is far better than a CPU at generating graphics.
AMD's compute division does better every year. But Nvidia's growth is outpacing AMD's growth in data center. So if AMD goes from 5 to 7 and Nvidia goes from 10 to 27, that results in the latter's market dominance despite AMD's growth.
My guess is someday, something else will overtake machine learning. It won't happen tomorrow, next week, or even next year. But it is bound to happen and if Nvidia is alert, they will adjust to the changing conditions. If they didn't it is doubtful they would be in business today.
And people in this discussion forum need to remember that whatever Nvidia and Intel do doesn't prevent AMD from assigning some engineers to research machine learning technologies in their R&D labs.
It's up to all businesses to adapt to the changing environment and customers' wishes. If you had a limited amount of flour and oven space and people will pay way more for your pie than your cookies, well, you'd make pie. If a farmer gets more from ambrosia cantaloupe than watermelon, he's going to plant cantaloupe instead. In the summer make lemonade, in the winter make hot chocolate.
This is not rocket science people, these are basic business principles on how to run a company that will last. This is nothing new, it goes back to the first time humans started producing things.
I could be totally off base and they pull the pin on the gaming segment all together, but I suspect that yeah, it's about more than just margins and shareholders, there's a heap of value in perceptions, and I think there's ego at stake there too.
I hope this is only the beginning and not the end of a chapter in the history of AI. And the rest of the market parties will catch up in time.
The handing on a platter part is they could have sold chips to MS at a loss for a year or two, it would have bankrupted AMD at the time. They gave AMD a lifeline instead.
Their later actions also support my interpretation: they opted to supply Nintendo a few years later and that's hardly likely to be a high margin business. Moreover, Nvidia of 2010-2013 wasn't Intel; they didn't have any cause to fear an anti-trust case and would have crushed AMD if they could have. That they didn't succeed in winning the Sony and Microsoft consoles was down to two things: lackluster SOCs, and more importantly, bad relationships with both Microsoft and Sony.
Nvidia Tegra 4 results below: notice that it is slower than the A4-5000 despite a nearly 30% higher clock (1.9 vs 1.5 GHz)
1. If Nvidia couldn't compete, how come AMD didn't crush them in the SoC market?
2. Without AMD, Nvidia would have become a GPU monopoly, that's what they were avoiding.
We are talking about 2013 not 2023. How would Nvidia have been investigated for anti-trust when AMD was still at 40% market share in discrete GPUs?