Register     Login Language: Chinese line English
padding: 100px 0px; text-align: center;">

X-trader NEWS

Open your markets potential

One AI Server Consumes 15 Times More Capacitors: MLCC, Substrates, and TCB Lead the Supply Chain Boom

News

One AI Server Consumes 15 Times More Capacitors: MLCC, Substrates, and TCB Lead the Supply Chain Boom


Due to the extremely high energy consumption of AI servers, their usage of MLCC (Multi-Layer Ceramic Capacitors) is as high as 15 times that of traditional servers, directly driving an explosive growth in demand for related core components. Currently, the production line utilization rates of MLCC, advanced substrates, and TCB (Thermal Compression Bonding) equipment all exceed 90%. The tight supply-demand pattern has enabled relevant suppliers to grasp core pricing power and launch product price increases. Among them, Asian suppliers such as Samsung Electro-Mechanics, Murata Manufacturing, and Unimicron Technology have become the biggest beneficiaries of this round of supply chain boom.
The wave of AI infrastructure construction is continuously penetrating into the depth of the supply chain, and a group of previously little-known component manufacturers are accelerating into the sight of investors, ushering in new development opportunities.
With the continuous expansion of global AI computing power demand, the consumption scale of various electronic components by AI servers is far exceeding that of traditional servers, and the usage gap of some core components is even as high as dozens of times. According to Young Jae Lee, Senior Investment Manager at Pictet Asset Management, the MLCC consumption of a single AI server is 10 to 15 times that of a standard traditional server, and about 30 times that of a smartphone, showing a particularly significant usage gap.
The sudden surge in demand is continuously driving up the prices of related products. Samsung Electro-Mechanics stated this week that it is considering raising the price of MLCC products by up to 10%. Relevant analysis by Goldman Sachs pointed out that the market demand for MLCC will grow by 4.3 times between 2025 and 2030, and at the same time, it raised its price increase expectation to 0-5%, with the industry prosperity continuing to climb.
The capital market has taken the lead in responding to this industry trend. Substrate manufacturers Unimicron Technology and Ibiden have seen their stock prices soar by about 770% and 530% respectively in the past 12 months, with significant gains; MLCC industry leaders Samsung Electro-Mechanics and Murata Manufacturing both hit historical highs in their stock prices this month; Hanmi Semiconductor, a leading enterprise in the field of Thermal Compression Bonding (TCB), also touched a historical high in its stock price, with increasing market attention.

Three Key Component Categories: Synergistic Efforts from Capacitors to Substrates and Bonding Processes

In this round of AI supply chain boom, investors' focus is concentrated on three core component categories, whose market performance directly determines the prosperity of the supply chain.
The core function of Multi-Layer Ceramic Capacitors (MLCC) is to regulate the power distribution inside the electronic system, which is an indispensable basic core component to ensure the stable power supply of AI servers. Advanced chip substrates act as the connection link between semiconductors and other hardware, and are important carriers for the normal operation of chips. Kieron Poon, Director of Asian Equity Investments at Aberdeen Investments, explained with an vivid metaphor: "Imagine a printed circuit board as a dining table, the plates on the table are substrates, and the food on the plates is chips." Thermal Compression Bonding (TCB), on the other hand, is a key packaging process that precisely integrates the above-mentioned various components, which is crucial for the implementation of advanced packaging technology and directly affects the operating efficiency of AI servers.
The core logic driving the explosive growth in demand for the above three types of components lies in the high energy consumption attribute of AI infrastructure construction. The power consumption of AI servers is far higher than that of traditional servers, and higher power output means that more electronic components are needed to manage and stabilize power supply. This demand transmission forms a cascading effect, thereby driving the demand growth of the entire supply chain.

Capacity Utilization Exceeds 90%, Upward Price Pressure Continues to Accumulate

Capacity constraints on the supply side have gradually emerged, becoming an important support for promoting product price increases. Simon Woo, Head of Korea Research at Bank of America Global Research, said that the production line utilization rates of MLCC and advanced substrates have both exceeded 90%, and the capacity is operating at a high level. "If the demand in the AI field increases slightly again, the capacity available for traditional uses will shrink sharply," he added.
The expectation of price increases for related components is also continuing to heat up. Takayuki Naito, Analyst at Citigroup, pointed out that the current market expectation of price increases for various core components such as MLCC, aluminum electrolytic capacitors, and packaging substrates is constantly expanding, and this expectation will form a strong support for Japanese manufacturers such as Murata Manufacturing and Taiyo Yuden. It is expected that the two companies will adopt a more active pricing strategy to further release the price increase space. Analysts at JPMorgan Chase also raised their target prices for Murata Manufacturing and Taiyo Yuden last week, and predicted that the current tight supply-demand situation will continue for a long period of time, and the high industry prosperity is expected to continue.
From the perspective of the industry competition pattern, the market supply of the above core components is highly concentrated, and the vast majority of relevant production enterprises are concentrated in the Asian region, mainly distributed in three countries: South Korea, Japan, and China. At the same time, with the continuous expansion of global AI infrastructure investment, the customer group of related components is also expanding. Kieron Poon from Aberdeen believes that this supply-demand pattern means that the pricing power of core components "will undoubtedly remain in the hands of suppliers", and the bargaining power of suppliers will continue to improve.
The spillover effect of this market boom is also spreading to a broader AI ecosystem. Optical component manufacturers have also been involved in this round of upward wave. The core reason is that with the continuous popularization of AI applications, the demand for bandwidth in data centers is becoming increasingly strong, and investors have begun to pay closer attention to the important role played by optical technology in it, driving the demand for optical components to increase.

Risk Warning and Disclaimer

The market is risky, and investment needs to be cautious. This article does not constitute personal investment advice, nor does it take into account the special investment objectives, financial situation or needs of individual users. Users should consider whether any opinions, views or conclusions in this article are in line with their specific situation. Investment based on this is at the user’s own risk.

CATEGORIES

CONTACT US

Contact: Sarah

Phone: +1 6269975768

Tel: +1 6269975768

Email: xttrader777@gmail.com

Add: 250 Consumers Rd, Toronto, ON M2J 4V6, Canada

Scan the qr codeClose
the qr code