July 8, 2025 09:00 PM GMT Asia Technology | Asia Pacific AI ASIC: Reconciling Trainium2 chip shipments We attribute the AWS Trainium2/2.5 chip shipments mismatch (between semis and systems) to unstable PCB yield rates. We expect chip shipments to reach around 1.1mn in 2025. Reason for this report: We conducted follow-up research on AI Supply Chain: AI ASIC dynamics: Trainium and TPU, as some investors expressed confusion about the wide range of AWS Trainium2/2.5 chip shipment assumptions, from 1mn to 2mn un ...

June 30, 2025 09:00 PM GMT Global Technology AI Supply Chain: Semis Field Trip – Key Takeaways TSMC's 2026 CoWoS capacity is growing > 30% Y/Y, psoitive for both the Nvidia and AI ASIC supply chains. On the other hand, B30 GPU shipments to China will be a wild card for China's AI capex. Over the past week, our global semis team met with ~20 corporate and AI industry experts in the AI supply chain in Taiwan and Beijing. We also met investors there to get a sense of market sentiment. In general, investors are ...

China's AI application/inference demand is strong, but hardware supply is the bottleneck: Chinese AI developers (covered by our Internet analyst for China, Gary Yu) in Beijing are aware of NVIDIA's B30 (RTX 6000 Pro) chips, though no official purchase orders with NVIDIA have been confirmed. However, our checks with Taiwan's supply chain indicate foundry wafer orders for this chip totaling 2mn units in 2H25, and annual run-rate is around 5mn units. If B30 cannot be shipped to China, some Chinese AI developer ...

◼ AI ASIC 趋势已明朗，关注数据中心铜连接市场。 从目前已获知的 CSP ASIC 服务器方案来看，使用铜缆进行短距互 连已成趋势。根据 Fibermall 报告，AWS 今年有望采购 150 万颗自研芯 片，大部分使用AEC互连，尽管Trainium2的计算能力低于英伟达H100， 但可以使用 400G AEC，随着 Trainium3 于年底推出，800G AEC 的需求 有望得到增加；虽然微软在使用 AEC 铜缆方面较 AWS 缓慢，但目前已 经开始使用 AEC 构建 AI 网络；谷歌 TPU 互连目前主要使用无源铜缆 DAC，随着速度的提升，可能会转向 AEC。尽管目前还没有 Meta 使用 AEC 铜缆的明确信息，但在其 Minerva 机柜系统中，使用了 2 个可拆卸 电缆背板盒来连接计算托盘以及网络托盘。该电缆背板盒使用了 112G PAM4 连接器和电缆，每个盒子包括 4 个电缆组，每个电缆组包含 384 对不同电缆线，在 8 个 MTIA 托盘，6 个网络托盘和 1 个机柜管理系统 托盘间形成了 1536 对电缆网络。头部 CSP 厂商之外，X.AI 也对 AEC 表现出了大量需求， ...

Core Viewpoint - Marvell Technology (MRVL.US) has seen a significant stock price increase due to positive evaluations from top Wall Street analysts regarding its customized AI ASIC chip activities and potential market announcements [1][3] Group 1: Market Opportunities - Analysts from Evercore ISI predict that the new AI ASIC chip designs could ramp up quickly between 2026 and 2027, indicating strong future demand [1] - Marvell expects each customized AI chip design win to generate billions in lifecycle revenue within 1.5 to 2 years, while each XPU Attach win could contribute hundreds of millions within 2 to 4 years [2] - The total addressable market (TAM) for customized data center chips has been raised to $94 billion, a 26% increase from last year's AI activities [3] Group 2: Financial Projections - Marvell has raised its financial targets, with analysts noting that the potential earnings per share could reach $8 by 2028, exceeding Wall Street estimates by 60% [4] - The company aims to capture at least 20% of the TAM, with over 50% of its data center revenue expected to come from AI ASIC-related demands [3][5] Group 3: Competitive Landscape - Broadcom (AVGO.US) is identified as the long-term beneficiary of Marvell's AI activities, holding a dominant market share of approximately 60% in the AI ASIC sector, while Marvell holds 13% to 15% [6][7] - The AI ASIC market is expected to grow significantly, with major tech companies like Google, Microsoft, and Amazon investing heavily in AI ASIC chips, indicating a shift in market dynamics away from GPU dominance [7]

Hutt表示，虽然英伟达的Blackwell芯片在性能上优于Trainium2，但AWS芯片在性价比上更具优势。 "Trainium3将于今年推出，其性能是Trainium2的两倍，并将在节能方面再提升50%。"他说。 AWSAnnapurnaLabs工程总监RamiSinno表示，目前市场对这些芯片的需求已经超过供应。 "我们的供应能力非常强大，但我们开发的每一项服务背后都已经有客户在排队等着用了。"他说。 亚马逊云服务（AWS）即将宣布其Graviton 4芯片的一项升级：将网络带宽提升至每秒600Gbps，这也 是AWS称之为"公有云中最高规格"的配置。 AWS的杰出工程师AliSaidi将这一速度比作"一台机器每秒可读取100张音乐CD"。 Graviton 4是一款中央处理器（CPU），由亚马逊(AMZN.O)位于德州奥斯汀的Annapurna Labs设计开 发，是亚马逊定制芯片战略的一部分，用于与英特尔(INTC.O)和AMD(AMD.O)等传统半导体巨头竞 争。 但真正的战场在于人工智能基础设施领域，与英伟达的正面对决。 在AWS去年12月的re:Invent2024大会上，亚马逊宣布了Pro ...

智通财经APP获悉，亚马逊(AMZN.US)旗下云计算服务平台AWS将宣布对其Graviton4芯片进行更新， 该芯片的网络带宽达到每秒600千兆比特，该公司称这是公共云中的最高带宽。 AWS工程师Ali Saidi将这种速度比作一台机器每秒读取100张音乐CD。 Graviton4是一款中央处理器(CPU)，是亚马逊位于德克萨斯州奥斯汀的Annapurna实验室推出的众多芯 片产品之一。这款芯片是该公司定制战略的一大胜利，使其能够与英特尔(INTC.US)和AMD(AMD.US) 等传统半导体厂商展开竞争。 但真正的战斗是在人工智能基础设施领域挑战英伟达(NVDA.US)。 在去年12月举行的AWS re:Invent 2024大会上，该公司宣布了为初创公司Anthropic打造的人工智能超级 计算机Project Rainier。AWS已经投入80亿美元支持Anthropic。 根据AWS的说法，Anthropic的Claude Opus 4人工智能模型是在Trainium2 GPU上启动的，而Project Rainier则由超过50万块芯片驱动——这笔订单传统上会落入英伟达手中。 Hutt表示，虽然英 ...