AI server 結果共84筆

Taiwan’s thermal management suppliers closed May 2026 with some of the strongest growth numbers in the AI supply chain. The results landed a week after Computex 2026, where chips with thermal design power above 1,000 watts and racks headed toward one megawatt made one thing plain: cooling has moved from the periphery of server design to the center of its economics.

Taiwan’s server manufacturers closed out May 2026 with another round of record monthly revenues. The reports came days after Computex 2026 wrapped in Taipei on June 5, where Nvidia CEO Jensen Huang used his keynote to reaffirm the sheer scale of AI infrastructure demand.

The PC came to Computex 2026 as a guest in its own house. The keynotes belonged to server silicon, the biggest booths to rack builders, and the show’s center of gravity to the data center. Yet the week still produced the most consequential PC news in a decade, because the machine itself is being rebuilt around AI agents at the exact moment its economics are being rewritten by AI’s appetite for memory.

Some of the biggest crowds at Computex 2026 were not gathered around AI server racks but the liquid cooling and power delivery systems that keep them running. In a clear sign of how the AI boom is changing, the booths of Delta, Lite-On, Auras, Jentech, and other companies in the segment saw lines stretching through the aisles. For years, cooling pipes and power supplies were boring background parts that only mattered when something broke. Now they have become one of the main stories. As AI racks grow more powerful, delivering the electricity they demand and removing the heat they generate have become two of the toughest challenges in the data center.

While the silicon vendors captured the keynote headlines, Taiwan’s hardware manufacturers drew the biggest crowds on the show floor. Their expansive booths featured server racks, advanced cooling systems, robots, and complete computing platforms. The attention was well deserved. Taiwan assembled roughly 90 percent of the world’s AI servers in 2025, and total server shipments are forecast to rise 19.2 percent in 2026, marking the strongest growth in years. Global shipments in the second quarter alone are projected to exceed five million units for the first time. These manufacturers turned Computex into a vast showroom for the physical machines powering the AI boom.

At Computex 2026, the silicon industry did more than simply attend the show. It realigned. While crowds focused on server racks, dexterous robots, and high-speed interconnects, the deeper story appeared in the chip announcements and partnership deals. The old battle lines of the processor world are dissolving. Nvidia has entered the CPU market, Arm is building its own data-center processors, MediaTek is expanding from phones to AI servers, and even Intel and Nvidia have found reasons to cooperate. What emerged is a new map of the industry, shaped by the demands of agentic computing and held together by alliances as much as by competition. All of this continues to rely on Taiwan’s unmatched manufacturing ecosystem.

The crowds arrived early and never thinned. From the moment the gates opened each morning, the aisles of the Nangang Exhibition Center filled with a press of analysts, buyers, engineers, reporters, and tech enthusiasts that did not let up until the halls closed in the evening. Booths that in previous years drew a polite trickle were three deep in people leaning in to watch a robot arm sort parts or a coolant loop pulse through a transparent server rack. There was a charge in the air that trade shows rarely carry, the sense of being inside something while it is still happening rather than reading about it afterward. The banner overhead read ＂AI Together,＂ and for once the slogan matched the feeling on the floor.

Compal Electronics showcases its collaboration with Nvidia on Physical AI at Computex. The company shifts from PCs to AI servers, investing $500M in Texas and planning a new Taiwan facility. Compal focuses on AI infrastructure, 5G connectivity, and medical tech, with energy availability as a key factor.

MiTAC unveiled AI infrastructure products at Computex Taipei, including liquid-cooled racks and diamond-cooled servers, enhancing efficiency and reducing costs.

Pegatron joins Nvidia’s next-generation Vera Rubin server supply chain, Chairman Tung Tzu-hsien announces at Computex Taipei. Company unveils AI products.

Start with AI infrastructure. Nvidia CEO Jensen Huang announced in Taipei last week that the company will invest roughly US$150 billion a year in Taiwan, a tenfold increase on five years ago, and described the island as “the epicenter of the AI revolution.” It is hard to argue with the numbers. Each Vera Rubin AI server, Nvidia’s next-generation flagship platform ramping in the second half of this year, contains nearly two million components and needs between 100 and 150 Taiwan ecosystem partners to assemble. Huang has called it “potentially the largest product rollout Taiwan’s electronics industry has ever seen.”

The restaurant was the Brick Kiln, a good but unflashy Taipei spot far from any conference hall. The host stood on a stool to be seen. The guest list ran to more than thirty CEOs, between them representing close to a trillion US dollars in market capitalization and almost every link in the global AI hardware supply chain. Jensen Huang toasted, thanked, moved from table to table, told an affectionate joke about a portrait he said looked more like TSMC’s C.C. Wei than himself, and at the end of the evening walked out into the street to hand sweets to the families and children who had gathered hoping for a glimpse. ＂Chemistry＂ was the word the Nvidia blog post used to describe the event. The casualness was the point.

Most of the attention in the AI server boom goes to the companies building the servers. Delta Electronics makes the power and thermal layer they all depend on.The Taoyuan-based firm heads into Computex 2026 with its strongest first-quarter results ever and near-record April revenue. Chairman Ping Cheng recently told investors that customer demand is “very, very large” and current capacity “absolutely insufficient” – language Delta has rarely used in any prior cycle. Its booth theme, “Superior Efficiency, Shaping Sustainable AI,” reflects two decades of quiet mastery over the technologies now most critical to AI: high-efficiency power conversion, advanced cooling, and their tight integration.

How did Taiwan become, in Jensen Huang’s words, the epicenter of the AI revolution? The first major step came following the launch of the DGX-1.

Foxconn will arrive at Computex 2026 in its best operating position ever. Fresh off record first-quarter results and an even more impressive April, the world’s largest contract electronics maker is riding robust AI server demand, a rapid Nvidia Vera Rubin ramp, and accelerating progress in EVs and robotics. Chairman Young Liu has designated 2026 a “strong growth” year, the company’s highest internal rating, and the financials are delivering.

Taiwan’s printed circuit board (PCB) makers turned in one of the strongest April results of any component segment, with several manufacturers reporting record monthly revenue on the back of AI server demand.

Taiwan’s AI server makers delivered another month of bumper year-on-year growth in April, as the global AI infrastructure build-out continued at full pace.

The April 2026 monthly revenue reports filed on the Taiwan Stock Exchange confirmed what the first quarter had already strongly implied: the AI infrastructure cycle is no longer a story about TSMC and a handful of server assemblers. It is now spread across the full hardware stack, from copper-clad laminates and ABF substrates to liquid-cooling modules, server rails and memory.

Computex 2026 opens on June 2 in Taipei under the banner “AI Together,” and for Taiwan’s server makers the timing is close to perfect. The first quarter delivered record results across the entire group. Hyperscaler orders for Nvidia’s next-generation GB300 rack systems are accelerating into the second half of the year. And a new category of compute that the industry has started calling Physical AI, is opening a fresh growth lane alongside the familiar workloads of training and inference.

When the AI server industry began standardizing on liquid cooling in 2025, Sunonwealth Electric Machine Industry, the Kaohsiung-based fan and motor specialist better known as Sunon, looked like the cooling supply chain’s most exposed name. A year later, that reading has aged badly. Founded in 1980 and listed on the Taiwan Stock Exchange under ticker 2421, Sunon has used its fan and motor heritage as a launchpad rather than a millstone, building out a credible liquid-cooling portfolio while its legacy business continues to grow.

Computex has long been a PC-centric event. It is the venue where Asus, Acer, MSI, Gigabyte, ASRock and other Taiwan manufacturers traditionally unveil their latest motherboards, laptops and desktop PCs, chassis, graphics cards and peripherals to global customers and the tech media.

After nearly four decades of operating quietly in the background, Jentech Precision Industrial has become one of the most important names in Taiwan’s AI cooling supply chain. Headquartered in the Guishan industrial district of Taoyuan and listed on the Taiwan Stock Exchange under ticker 3653, the company was founded in 1987 and offers a textbook example of how a precision-component specialist can adapt to the demands of the AI era.

Taiwan’s server orders from China are declining as Beijing accelerates AI self-sufficiency. U.S. hyperscalers continue avoiding Chinese suppliers, MIC reports.

One of the least remarked features of Taiwan’s AI hardware boom is that it has not been led by the kind of companies the global business press usually celebrates. There are no garage startups here, no twenty-something founders, no venture capital origin myths. The firms riding the AI infrastructure wave, supplying cold plates to GB300 racks and vapor chambers to hyperscaler servers, are for the most part middle-aged industrial companies that have been grinding away in the component trenches for three decades or more.

For decades, Kaori Heat Treatment quietly built its reputation making brazed plate heat exchangers for HVAC, refrigeration, and industrial applications. Today, the Taiwan company is one of the fastest-growing players in the AI infrastructure supply chain and its transformation is just getting started.