DIGITIMES ICT: Server, IPC, cloud computing, IoT news

Home Tech ICT Server, IPC, cloud computing, IoT

Tuesday 3 March 2026

Yageo subsidiary Kemet hikes tantalum capacitor prices for third time as strong AI server demand continues

As artificial intelligence (AI) server applications consume large amounts of tantalum polymer capacitors, the supply-demand gap in the market continues to widen. Kemet, a subsidiary of passive component giant Yageo, has issued another price increase notice, with the new pricing set to take effect on April 1.

Tuesday 3 March 2026

MediaTek highlights 6G, Wi-Fi 8, and AI chip UCIe tech at MWC 2026

At the 2026 Mobile World Congress (MWC), MediaTek showcased a series of breakthrough technologies under the theme "AI for Life: From Edge to Cloud," emphasizing its leadership in chips and AI. The company focused on 6G communications, next-generation home broadband platforms, automotive connectivity, mobile AI, and data center interconnect solutions.

Tuesday 3 March 2026

AI data centers redraw the power map: Driving 800V DC and solid-state transformers into the next battleground

As demand for computing power from AI large language models (LLMs) continues to surge, power density in data centers is rising in tandem, bringing the conversion losses of traditional power architectures to the forefront. The market expects 2026 to mark a critical transition period for the accelerated adoption of new technologies such as high-voltage DC power architectures and solid-state transformers (SSTs). Mastery of these power supply chain technologies is set to become a key barrier to entry in the next phase of market competition.

Tuesday 3 March 2026

Apple reportedly explores Google Cloud to power next-generation Siri

Apple is in discussions with Google about hosting a new version of Siri in Google's data centers, a move that could significantly deepen Apple's reliance on outside cloud infrastructure as it pushes to modernize its AI capabilities.

Tuesday 3 March 2026

Tesla's Optimus push deepens rivalry as Nvidia's ecosystem strategy reshapes global robotics competition

Tesla's pivot to free factory space for its humanoid robot Optimus by cutting Model S and Model X production lines highlights a broader shift in the robotics industry toward a structural competition between Tesla's closed-off approach and an Nvidia-led alliance connecting automakers and robotics firms. The move raises questions about risks tied to vertical integration and the alternative offered by platform-driven collaboration.

Tuesday 3 March 2026

Nvidia invests in Lumentum to advance AI optics technology

Nvidia has announced a strategic partnership with Lumentum Holdings, accompanied by a US$2 billion investment aimed at boosting US-based manufacturing and R&D for advanced optics technologies. The deal includes multiyear purchase commitments and access rights to next-generation laser components, with a focus on accelerating development for AI data centers.

Tuesday 3 March 2026

Ablecom ABLERACK targets high-density AI deployments with seismic-tested L11 cabinet

Ablecom announced its ABLERACK AI server rack solution, positioning it in the L11 domain and offering an integrated design for manufacturing services that combine energy efficiency, power distribution, and modular maintenance in a single cabinet.

Tuesday 3 March 2026

Advantech's special dividend plan signals management confidence in long-term outlook

Industrial PC provider Advantech said its board approved a limited three-year special dividend program and shifted to an all-cash dividend policy, increasing the regular cash payout ratio and planning extra annual cash payments of NT$2 (US$0.06) per share from 2026 through 2028, totaling over NT$5.2 billion.

Monday 2 March 2026

Middle East escalation disrupts AWS data center, Apple retail, and Robotaxi services

US and Israeli airstrikes on Iran triggered retaliatory missile attacks on US bases in Bahrain and Qatar, with spillover disruptions hitting cloud infrastructure, logistics hubs, and commercial operations across the Middle East.

Monday 2 March 2026

AMD deal with Meta puts CPU firm as leading alternative to Nvidia in AI compute

AMD secured a five-year agreement with Meta Platforms Inc. to supply up to 6GW of AI chips, a deal Meta said could be worth more than US$100 billion, and which industry observers say establishes AMD as the primary alternative to Nvidia in large-scale AI infrastructure.

Monday 2 March 2026

Memory price surge spreads to servers; Wiwynn flags sharp post-holiday cost jump

Memory price increases are beginning to affect the server market. Wiwynn said that in 2025, server customers and memory suppliers were protected by long-term pricing agreements, but price growth accelerated sharply after the 2026 Lunar New Year. Customer demand remains unchanged. Higher memory costs will pressure ODM gross margins, while final profit and earnings per share (EPS) are expected to remain stable.

Monday 2 March 2026

Trade tensions deepen Nvidia and TSMC's strategic grip on AI infrastructure

As expected, Nvidia delivered another strong earnings beat for the fourth quarter of its fiscal 2026, reinforcing the view that global demand for AI computing remains resilient despite mounting geopolitical pressures. For the fiscal year, the company posted more than US$120 billion in profit with a gross margin of 71.1%, underscoring the extraordinary profitability of AI infrastructure even as export controls and trade tensions intensified during the first year of US President Donald Trump's administration.

Monday 2 March 2026

Nvidia confirms liquid cooling as standard, boosting supply chain revenue outlook

Nvidia delivered a strong fourth quarter fiscal year 2026 earnings report and issued an optimistic forecast, driven by robust sales of its GB architecture AI servers. Notably, liquid cooling paired with the GB architecture has officially become a standard feature. Nvidia stressed that liquid cooling no longer poses a shipment bottleneck, dispelling earlier industry concerns sparked by CEO Jensen Huang's CES remarks.

Monday 2 March 2026

Analysis: AMD bets on AI surge in 2H26 with OpenAI and Meta ecosystem pact

AMD is stepping up efforts to expand its position in the artificial intelligence semiconductor market, combining long-term customer agreements with equity-linked incentives as it seeks to narrow the gap with rival Nvidia.

Monday 2 March 2026

Hiwin local production positioned as key supplier for South Korea semiconductor and AI-driven equipment upgrades

Hiwin Technologies has spent more than a decade building direct manufacturing and service capacity in South Korea, evolving from component sales to integrated mechatronic solutions and becoming a preferred supplier as semiconductor upgrades and AI-related investment accelerate.

Monday 2 March 2026

Commentary: Nvidia sparks silicon photonics race, yet copper still anchors data centers?

As data center computing demand expands, high-speed transmission architectures are under pressure to upgrade. Scaling high-performance computing platforms is exposing bandwidth and power consumption as core bottlenecks in data transmission and switching, elevating the role of optical interconnects and silicon photonics (SiPh). With support from AI chip leader Nvidia, the notion of optics gradually replacing copper has moved to the center of industry debate.

Monday 2 March 2026

Insight: Memory shortage tightens grip on smartphone market as prices near tipping point

The memory market is no longer just a component story — it is becoming a fault line running through the entire tech industry. As AI infrastructure buildout accelerates, cloud and data-center operators are consuming DRAM and NAND at a pace that is crowding out smartphone makers, distorting foundry economics, and forcing chipmakers to rethink how they secure supply. The consequences are rippling from factory floors in Asia to boardrooms in Silicon Valley.

Monday 2 March 2026

Google brings Intrinsic in-house to accelerate physical AI development

Google has announced that robotics software company Intrinsic will join its operations as a distinct unit within the company, a move aimed at accelerating the deployment of artificial intelligence in physical systems such as industrial robotics. The integration reflects Google's effort to extend AI beyond digital applications into real-world environments and scale practical physical automation solutions.

Monday 2 March 2026

Nvidia's Vera Rubin platform faces HBM4, cooling, and software hurdles ahead of ramp

Nvidia's next-generation Vera Rubin platform has moved from public unveiling to early customer sampling, with the company projecting a broader production ramp later this year. Both the company and its partners, however, face a complex array of engineering, supply-chain, and data center infrastructure challenges before Rubin can displace prior architectures as the industry standard for large-scale artificial intelligence.

Sunday 1 March 2026

Taiwan's Fitipower eyes stronger year as AI and edge chips gain traction

Fitipower, a Taiwanese driver IC and semiconductor supplier, offered a cautiously optimistic outlook for 2026 during its February 25 investor briefing. Chairman Young Lin said the company expects full-year performance to surpass 2025, with growth increasingly coming from non-display IC applications, including AI SoCs and edge computing solutions.

Sunday 1 March 2026

BizLink advances modular integration to boost data center interconnect growth

Despite no signs of a slowdown in AI infrastructure investments by global cloud service providers (CSPs), debates about a potential AI bubble persist. BizLink CEO Felix Teng emphasized that the focus should shift from AI itself to the underlying demand for computing power and the broader infrastructure investment cycle.

Friday 27 February 2026

Wiwynn's AI products exceed 50% of sales in 2025, revenue and profit double

US cloud service providers (CSP) continue to invest heavily in artificial intelligence (AI), with Nvidia's latest fourth-quarter fiscal 2026 earnings report showing record-high revenue, highlighting ongoing strong demand for AI infrastructure. Benefiting from the AI boom, Wiwynn reported consolidated revenue of NT$950.66 billion (US$30.46 billion) in 2025, up 163.7% year over year, with AI products accounting for more than half of sales.

Friday 27 February 2026

Nobel laureate John Martinis explains quantum computing's industrial revolution from lab to fab

As AI dominance drives computing costs to unprecedented heights, a surprising truth is emerging at the front of quantum technology: the future of AI may depend not on replacing classical computers, but on a subtle, yet astronomically valuable, incremental improvement.

Thursday 26 February 2026

BizLink sees HPC and semiconductor revenue surpass 50% with AI data center upgrade wave

As generative AI drives rapid expansion in data centers and high-performance computing (HPC) infrastructure, the importance of interconnect technology is evolving from traditional signal transmission to power delivery and system integration. Connector and cable maker BizLink Holding has seen a significant shift in its business structure, with HPC and semiconductor segments already accounting for over 51% of total company revenue by the third quarter of 2025, reflecting growth momentum shifting toward AI infrastructure and semiconductor equipment sectors.

Thursday 26 February 2026

Nvidia constrained in China as local AI players strengthen market position

Nvidia's ability to sell high-performance AI chips in China has been sharply limited by US export licensing rules, which have restricted shipments of its H20 and H200 products. The controls have already forced Nvidia to take a US$4.5 billion inventory charge and have left the company uncertain about future revenue from China. Meanwhile, local competitors, some strengthened by recent IPOs, are expanding rapidly, potentially reshaping the global AI market.

More Server, IPC, cloud computing, IoT coverage