Author: Zhou Yuan / Wall Street Watch

Recently, Broadcom showcased its performance strength in AI technology and products in the fiscal year 2024 performance report released on December 12, shocking the Global market.

Due to the market acceptance of Broadcom's ASIC (Application-Specific Integrated Circuit), NVIDIA, a company producing general-purpose computing chips, saw its shares plunge nearly 7%, hitting a new low since October 18 during trading.

What exactly does Broadcom do?

Broadcom's Business primarily covers two major areas: semiconductor technology and infrastructure Software, demonstrating a strong and diversified characteristic.

In the fiscal year 2024, Broadcom's net income reached 51.6 billion USD, with a gross margin of 76.5%, showcasing remarkable profitability and operational efficiency.

Among them, the semiconductor business revenue was 30.1 billion USD, and infrastructure Software business revenue was 21.5 billion USD, a year-on-year increase of 196%. In the AI sector, Broadcom's performance was particularly outstanding, with AI business revenue reaching 12.2 billion USD, a year-on-year increase of 220%, significantly increasing its share in the semiconductor business, fully demonstrating its competitive strength in the AI era.

Broadcom's overall technical strength includes network chip technology, AI custom chip technology, and interconnection technology. Like NVIDIA, Broadcom also has a unique Software and ecological layout.

In the era of high-performance computing and AI, network chips are an important pillar for Broadcom.

As the largest network chip manufacturer in the Global market, Broadcom's Ethernet product line provides various solutions for AI clusters. For example, the Tomahawk 5 Ethernet Switch Chip in the Endpoint Scheduled system architecture has achieved an 80-fold bandwidth increase since 2010—from 640Gbps to 51.2Tbps in 2022—while also reducing energy consumption by over 90%, efficiently supporting data scheduling for small-scale AI clusters.

In the Switch Scheduled architecture, Ramon and Jericho3 - AI Chip can achieve multi-path interconnection, with Jericho3 - AI capable of connecting up to 32,000 GPUs. Each AI acceleration card provides 800Gbps of data bandwidth, effectively improving network performance.

In addition, Broadcom's Thor series network cards are continuously innovating: Thor 2 is the industry's first 400 gigabit Ethernet NIC device based on 5nm technology, supporting 16 PCI Express 5.0 lanes, capable of directly driving 5-meter Copper cables, while also supporting RoCE v2 RDMA technology. Broadcom is also developing higher-speed network card chips, continuously leading the development of network chip technology.

In the field of AI custom chip technology, Broadcom is the largest AI custom chip service provider in the Global market, holding about 70% of the $17.5 billion serviceable market.

Since entering the customized AI accelerator market in 2014, Broadcom has collaborated closely with customers to develop multiple generations of XPU accelerators.

Broadcom plans to launch the next generation XPU based on 3nm technology in the second half of 2025, accelerating high-performance computing development.

In terms of IP investment, Broadcom is very forward-looking, having invested approximately 3 billion USD in differentiated and innovative XPU IP, possessing a wide range of IP portfolios that include SerDes IP, AI-optimized NICs IP, cache IP, CPO IP software API, etc., providing crucial support for XPU.

At the same time, Broadcom has also made significant breakthroughs in packaging technology, with its 3.5D eXtreme Dimension system-level packaging (XDSiP) platform technology integrating a large number of chips and HBM stacks, showing significant advantages in interconnect density and power efficiency, effectively improving the efficiency of large-scale AI computing and reducing power consumption.

Interconnect technology and related products are also areas of technological advantage for Broadcom.

For example, in PCIe + Retimer technology, Broadcom has a strong capability in the PCIe technology field.

What is PCIe.

PCIe (Peripheral Component Interconnect Express), a high-speed peripheral component interconnect standard, is a high-speed serial computer expansion bus standard used to connect the computer's main board and various peripheral devices.

Since the launch of the first generation PCIe in 2003, Broadcom has cumulatively shipped over 1 billion PCIe ports. In 2018, the Retimer technology was introduced to solve long-distance signal transmission issues, becoming a standard in the subsequent development of PCIe. In March 2024, Broadcom launched the world's first 5nm PCIe Gen 5.0/CXL2.0 and PCIe Gen 6.0/CXL3.1 re-timer, providing important guarantees for achieving high-speed interconnections.

In the field of optical interconnects, Broadcom has long been in a leading position, especially with fruitful results in optical interconnect technologies such as VCSEL and EML high-linearity continuous wave lasers. These technologies play a key role in high-speed interconnections in AI and ML systems, with speeds iterating up to 200Gbps.

Currently, Broadcom ships over 50 million high-speed optical communication lasers annually.

Broadcom has launched the industry's first 51.2Tbps CPO Ethernet Switch, Bailly, which significantly reduces the running power consumption of optical interconnects and improves silicon area efficiency compared to pluggable transceivers. Its 5nm Sian 2 and similar products also feature robust electrical and optical interface capabilities.

Additionally, the 5nm Peregrine and 3nm Condor SerDes technology developed by Broadcom not only supports Copper cable connections but also natively supports CPO, which can be widely applied across multiple product platforms. This technology boasts bandwidth capabilities exceeding 100GB, while the best domestic counterparts only achieve 2GB.

Broadcom has also achieved remarkable results in infrastructure Software, with its revenue accounting for 41.6% of total revenue.

Through a series of strategic acquisitions, such as the acquisition of software giant VMware in 2023, Broadcom has built a relatively complete software business system, further enhancing its competitiveness in the Semiconductors technology and infrastructure Software market, providing customers with more comprehensive solutions and creating an efficient collaborative network and software ecosystem.

However, to be realistic, the integrity and ease of use of Broadcom's ASIC ecosystem are not on par with NVIDIA's CUDA; the software environment is relatively singular, mainly designed for specific purposes and algorithms, with a higher programming difficulty, and the development tools and software libraries are not abundant enough, requiring developers to spend more time on debugging and development, which limits its promotion in broader application scenarios.

In comparison, NVIDIA's GPU products have a fast iteration speed, keeping up with technological trends and market demands, with new products or performance improvements released every year.

In contrast, once Broadcom's ASIC chips are designed and manufactured, they are difficult to modify, with a longer design and manufacturing cycle, making them less suitable for the fast-changing AI market's frequent technological iterations, which may lead to a lag in technological advancement compared to NVIDIA.

The inability to make adjustments in technology and manufacturing as needed is the main reason why customized chips are more expensive than generic chips.

Moreover, customized chips have another issue: limited customer base.

Broadcom mainly focuses on providing customized chip services for large Technology companies, resulting in a relatively narrow customer base; NVIDIA's GPU products are highly versatile, meeting the needs of businesses of various sizes and applications, from small startups to large enterprises.

Once the major customers that Broadcom relies on make any technological shifts or business adjustments, Broadcom's performance will also be significantly negatively affected.

From this perspective, NVIDIA has a greater advantage in market coverage and customer diversity, allowing it to better respond to market fluctuations and the changing needs of different customers.

The lack of a full-stack solution is also Broadcom's shortcoming.

Broadcom mainly focuses on chip design and packaging, lacking a full-stack solution like NVIDIA: customers using Broadcom chips need to solve issues with network devices, Software, and other supporting infrastructure on their own, increasing the costs and complexity of use for customers.

Perhaps the market has fully recognized Broadcom's shortcomings when facing NVIDIA, which is why after the release of the fiscal year 2024 earnings report on December 12, yesterday (December 23) after the close of the U.S. stock market, NVIDIA regained its "lost ground" in stock price, overall slightly rising by 1.70%; while Broadcom's stock price fell by 6.84% from the peak on December 14 to $232.5 on December 23, but its total market value is still above 1 trillion dollars, standing at 1.09 trillion dollars.

In terms of future prospects, Broadcom is not only a leader in the Semiconductors Industry but also a key player in driving the development of AI Technology.

From network connectivity to AI computing power, from Semiconductors Hardware to Software infrastructure, Broadcom's technological puzzle is complete and powerful, providing a solid foundation for the development of the AI Business.

Of course, Broadcom also faces many challenges, such as increasing market competition, requiring continuous investment in research and development to maintain its CSI Leading Technology Index; at the same time, the rapid development of AI Technology requires Broadcom to innovate continuously to adapt to the ever-changing market demands.