This market research report was originally published at Yole Développement’s website. It is reprinted here with the permission of Yole Développement.
Data center processor diversification is ongoing with AI co-processing & DPU acceleration.
OUTLINE
-
The processor market for data center servers is expected to reach US$106 billion in 2027.
-
The cloud computing segment is growing as individuals and businesses use more and more cloud services.
-
Intel, Nvidia, and AMD: the three giant American companies own most of the processor market for data center servers. Together Intel and AMD own 90% of the CPU market, and together Nvidia and AMD have almost all of the GPU market
-
The giant cloud and technology companies are playing a growing role in this processor for the data center market.
The growth of AI is having a strong impact on server processor technology. First, extensions enabling accelerated AI inferences are being added to CPUs. Then, the growth of AI is strongly pushing the attachment rate of AI co-processors.
“The first devices to benefit from this demand are GPUs, which are massively used to train AI technologies. AI ASIC co-processors are also growing, both for inferences and training. In parallel, the rise of networking acceleration with SmartNIC and DPU is happening, mainly in hyperscale data centers”.
Adrien Sanchez
Technology and Market Analyst for Computing and Software, Yole Intelligence
With the growth of virtualization, ever more CPU cores are dedicated to networking and data management tasks. Adding a networking co-processor such as SmartNICs and DPUs offloads heavy networking tasks from CPUs. This is the start of a trend called data center disaggregation, which is putting these new DPU processors at the heart of data center networks to handle networking tasks, as well as storage and infrastructure management tasks. This demand for a new type of processing is joined by a need for ever more servers to support the many growing applications. Together, these trends strongly drive revenue from processors for data centers, which will reach around US$106 billion in 2027, at a CAGR21-27 of 22%.
Yole Intelligence has developed a dedicated report to provide a scenario for computing within the dynamics of the data center market and presents a comprehensive understanding of AI’s impact on the semiconductor industry. In its new Computing and AI for Data Center 2022 report, the company, part of Yole Group, delivers an in-depth overview of the ecosystem and its players. It offers vital technical insights and analyses into future technology trends and challenges of computing and AI for the data center industry.
Although x86-based server CPUs still largely dominate the market, the diversity of other available products is growing. Ever more Arm-based solutions are in the market, with products from companies such as Nvidia, Ampere Computing, Amazon Web Services, and Alibaba. Meanwhile, with the decreasing pace of Moore’s law, advanced packaging and new technologies are essential to increase processor performance. This is leading to many innovations, such as TSVs , chiplets, and 2.5D and 3D packaging. They are used to bring memory closer to the processor by stacking stacking memory dies on the interposer. In parallel, many innovations are bringing computing and memory together in another way; computing inside memory units, with computational storage and processing in memory, for example.
Three main types of players are developing processors for data centers:
- Large fabless companies are the first category, with Intel, Nvidia, and AMD owning most of the CPU and GPU market today. Each of these companies is currently consolidating its solution portfolio through internal development and multi-billion-dollar acquisitions to offer a comprehensive data center solution.
- Then cloud companies are playing a growing role, from big processor customers to processor designers. Many have started designing their own processors for their internal needs.
- The third category is startups dedicated to developing AI accelerators. In the last decade, many have been created and received significant funding.
The Chinese ecosystem is particularly dynamic, motivated by several Chinese policies that aim to stimulate the semiconductor industry’s development, with a specific focus on AI.
Yole Intelligence’s computing team invites you to follow this industry with news, interviews, and detailed analyses on www.yolegroup.com.
Stay tuned!
Acronyms
- CPU : Central Processing Units
- GPU : Graphics Processing Units
- AI : Artificial Intelligence
- ASIC : Application Specific Integrated Circuit
- DPU : Data Processing Units
- SmartNIC : Smart Network Interface Cards
- CAGR : Compound Annual Growth Rate
- HBM : High Bandwidth Memory
