According to the Korea Economic Daily. Samsung Electronics’ HBM3 and packaging services have passed AMD’s quality tests. The upcoming Instinct MI300 series AI chips from AMD are planned to incorporate Samsung’s HBM3 and packaging services. These chips, which combine central processing units (CPUs), graphics processing units (GPUs), and HBM3, are expected to be released in the fourth quarter of this year.

Samsung is noted as the sole provider capable of offering advanced packaging solutions and HBM products simultaneously. Originally considering TSMC’s advanced packaging services, AMD had to alter its plans due to capacity constraints.

The surge in demand for high-performance GPUs within the AI landscape benefits not only GPU manufacturers like NVIDIA and AMD, but also propels the development of HBM and advanced packaging.

In the backdrop of the AI trend, AIGC model training and inference require the deployment of AI servers. These servers typically require mid-to-high-end GPUs, with HBM penetration nearing 100% among these GPUs.

Presently, Samsung, SK Hynix, and Micron are the primary HBM manufacturers. According to the latest research by TrendForce, driven by the expansion efforts of these original manufacturers, the estimated annual growth rate of HBM supply in 2024 is projected to reach 105%.

In terms of competitive dynamics, SK Hynix leads with its HBM3 products, serving as the primary supplier for NVIDIA’s Server GPUs. Samsung, on the other hand, focuses on fulfilling orders from other cloud service providers. With added orders from customers, the gap in market share between Samsung and SK Hynix is expected to narrow significantly this year. The estimated HBM market share for both companies is about 95% for 2023 to 2024. However, variations in customer composition might lead to sequential variations in bit shipments.

In the realm of advanced packaging capacity, TSMC’s CoWoS packaging technology dominates as the main choice for AI server chip suppliers. Amidst strong demand for high-end AI chips and HBM, TrendForce estimates that TSMC’s CoWoS monthly capacity could reach 12K by the end of 2023.

With strong demand driven by NVIDIA’s A100 and H100 AI Server requirements, demand for CoWoS capacity is expected to rise by nearly 50% compared to the beginning of the year. Coupled with the growth in high-end AI chip demand from companies like AMD and Google, the latter half of the year could experience tighter CoWoS capacity. This robust demand is expected to continue into 2024, potentially leading to a 30-40% increase in advanced packaging capacity, contingent on equipment readiness.

(Photo credit: Samsung)

US-based CSPs have been establishing SMT production lines in Southeast Asia since late 2022 to mitigate geopolitical risks and supply chain disruptions. TrendForce reports that Taiwan-based server ODMs, including Quanta, Foxconn, Wistron (including Wiwynn), and Inventec, have set up production bases in countries like Thailand, Vietnam, and Malaysia. It’s projected that by 2023, the production capacity from these regions will account for 23%, and by 2026, it will approach 50%.

TrendForce reveals that Quanta, due to its geographical ties, has established several production lines in its Thai facilities centered around Google and Celestica, aiming for optimal positioning to foster customer loyalty. Meanwhile, Foxconn has renovated its existing facilities in Hanoi, Vietnam, and uses its Wisconsin plant to accommodate customer needs. Both Wistron and Wiwynn are progressively establishing assembly plants and SMT production lines in Malaysia. Inventec’s current strategy mirrors that of Quanta, with plans to build SMT production lines in Thailand by 2024 and commence server production in late 2024.

CSPs aim to control the core supply chain, AI server supply chain trends toward decentralization

TrendForce suggests that changes in the supply chain aren’t just about circumventing geopolitical risks—equally vital is increased control over key high-cost components, including CPUs, GPUs, and other critical materials. With rising demand for next-generation AI and Large Language Models, supply chain stockpiling grows each quarter. Accompanied by a surge in demand in 1H23, CSPs will become especially cautious in their supply chain management.

Google, with its in-house developed TPU machines, possesses both the core R&D and supply chain leadership. Moreover, its production stronghold primarily revolves around its own manufacturing sites in Thailand. However, Google still relies on cooperative ODMs for human resource allocation and production scheduling, while managing other materials internally. To avoid disruptions in the supply chain, companies like Microsoft, Meta, and AWS are not only aiming for flexibility in supply chain management but are also integrating system integrators into ODM production. This approach allows for more dispersed and meticulous coordination and execution of projects.

Initially, Meta heavily relied on direct purchases of complete server systems, with Intel’s Habana system being one of the first to be integrated into Meta’s infrastructure. This made sense since the CPU for their web-type servers were often semi-custom versions from Intel. Based on system optimization levels, Meta found Habana to be the most direct and seamless solution. Notably, it was only last year that Meta began to delegate parts of its Metaverse project to ODMs. This year, as part of its push into generative AI, Meta has also started adopting NVIDIA’s solutions extensively.

DRAM Spot Market
TrendForce has found that in the past week, the supply of used DDR4 chips (that were stripped from decommissioned server DRAM modules) has started to tighten. Therefore, products in the lower price range have also experienced small price hikes. However, the biggest issue still lies with the demand situation. The spot trading of chips and modules will remain tepid as long as buyers lack the willingness to make significant procurements. The average price of the mainstream chips (DDR4 1Gx8 2666MT/s) fell by 0.21% from US$1.460 last week to US$1.457 this week.

NAND Flash Spot Market
The spot market is seen with fluctuations of prices among packaged dies at different capacities, though the tendency has not continued regardless. With suppliers constantly increasing their quotations, buyers are still at the wait-and-see end and have yet to fully turn aggressive in the stocking. Wafer prices, after several consecutive weeks of slow increment, are now drawing near the low market price interval. 512Gb TLC wafer spots have risen by 5.49% this week, arriving at US$1.519.

As reported by TechNews, a media partner of TrendForce, Southeast Asia and India, equipped with the advantages of demographic dividends, strategic geographic positioning, manufacturing capabilities, and rapidly growing economic markets, have undoubtedly emerged as the preferred destinations for the technology industry amidst the global supply chain transition prompted by geopolitical factors.

As supply chains actively seek production bases beyond China and governments introduce incentive programs and policy restrictions for localized supply, various Southeast Asian countries have become key hubs for different sectors. Vietnam has become a focal point for consumer electronics manufacturing such as laptops, watches, and headphones, while Thailand has become a preferred choice for automotive-related supply chains. Thailand and Malaysia host assembly bases for servers, and India is set to become a crucial hub for mobile phone production.

Apart from the movement of end-product assembling, the shift in the semiconductor supply chain has also garnered attention. With TSMC, Samsung, and Intel relocating wafer fabrication plants to the United States, Europe, and other regions, a significant cluster of semiconductor backend testing and packaging has been forming in Malaysia.

What Advantages Does Malaysia Offer to Attract Multinational Semiconductor Companies’ Investment, and What Is the Current Industry Landscape?

Firstly, Malaysia boasts higher education standards than neighboring countries. Among ASEAN nations, only Singapore and Malaysia employ the British legal system, providing a competitive edge for many companies’ location choices. Secondly, in terms of language proficiency, Malaysian citizens predominantly use English, Mandarin, and Malay, facilitating smooth communication with global enterprises.

Thirdly, Malaysia is home to two major ports—Port Klang and Port of Tanjung Pelepas—both ranked among the world’s top 15 ports, with substantial container handling capacity and global reach.

Lastly, the state of Penang stands as a semiconductor hub for Malaysia, having nurtured the semiconductor industry for several decades and holding a technological lead. Often referred to as the “Silicon Valley of the East,” Penang has primarily focused on producing chips for electronics, computers, and mobile phones. However, with the growing adoption of electric vehicles, the demand for automotive chips has surged. Concurrently, the green energy trend has propelled the need for solar panels and renewable energy sources. This optimistic outlook for the semiconductor industry has once again attracted numerous companies to establish facilities and expand production capacity.

Current State of Malaysia’s Semiconductor Industry

Looking at the recent dynamics of corporations over the past two years, the trend is evident that Malaysia is evolving into a center for semiconductor backend testing and packaging. Major global players have announced plans to establish or expand operations in Penang. Intel, for example, announced a $6.46 billion investment in Malaysia in 2021, focusing on advanced packaging capabilities in Penang and Kedah.

Texas Instruments declared its intent to construct semiconductor testing and packaging plants in Kuala Lumpur and Malacca, with a total investment of up to $2.7 billion. Infineon is investing $5.45 billion to expand existing facilities, producing silicon carbide and entering the electric vehicle sector. Bosch Group is investing $358 million in stages to strengthen its semiconductor supply chain position in Penang. ASE Technology Holding, also began construction on a new testing facility in Penang at the end of last year.

With the influx of semiconductor giants, Malaysia’s position in the semiconductor industry has become increasingly critical. The distinct production base trends, aligned with the strengths of various Southeast Asian countries, have become clear. The restructuring of supply chains and the transformation of production centers undoubtedly remain the focus and challenge for global companies.

(Photo credit: ASE)

Dell, a major server brand, placed a substantial order for AI servers just before NVIDIA’s Q2 financial report. This move is reshaping Taiwan’s supply chain dynamics, favoring companies like Wistron and Lite-On.

Dell is aggressively entering the AI server market, ordering NVIDIA’s top-tier H100 chips and components. The order’s value this year is estimated in hundreds of billions of Taiwanese dollars, projected to double in the next year. Wistron and Lite-On are poised to benefit, securing vital assembly and power supply orders. EMC and Chenbro are also joining the supply chain.

Dell’s AI server order, which includes assembly (including complete machines, motherboards, GPU boards, etc.) and power supply components, stands out with its staggering value. The competition was most intense in the assembly sector, ultimately won by Wistron. In the power supply domain, industry leaders like Delta, Lite-On, secured a notable share, with Lite-On emerging as a winner, sparking significant industry discussions.

According to Dell’s supply chain data, AI server inventory will reach 20,000 units this year and increase next year. The inventory primarily features the highest-end H100 chips from NVIDIA, with a few units integrating the A100 chips. With each H100 unit priced at $300,000 and A100 units exceeding $100,000, even with a seemingly modest 20,000 units, the total value remains in the billions of New Taiwan Dollars.

Wistron is a standout winner in Dell’s AI server assembly order, including complete machines, motherboards, and GPU boards. Wistron has existing H100 server orders and will supply new B100 baseboard orders. Their AI server baseboard plant in Hsinchu, Taiwan will expand by Q3 this year. Wistron anticipates year-round growth in the AI server business.