Workflow
Tomahawk 5
icon
Search documents
博通,悄然称霸
半导体行业观察· 2025-06-28 02:21
Core Viewpoint - The article emphasizes the importance of interconnect architecture in AI infrastructure, highlighting that while GPUs are crucial, the ability to train and run large models relies heavily on effective interconnect systems [1]. Group 1: Interconnect Architecture - Interconnect architecture encompasses various levels, including chip-to-chip communication within packages and system-level networks that support thousands of accelerators [1]. - Nvidia's dominance in the industry is attributed to its expertise in developing and integrating these interconnect architectures [1]. - Broadcom has been quietly advancing various technologies related to interconnect architecture, including Ethernet architectures for large-scale expansion and internal chip interconnect technologies [1][3]. Group 2: Ethernet Switch Technology - Broadcom has introduced high-capacity switches, such as the 51.2Tbps Tomahawk 5 and the recently launched 102.4Tbps Tomahawk 6, which can significantly reduce the number of switches needed for large GPU clusters [3]. - The number of switches required decreases as the switch's port count increases, allowing for more efficient connections among GPUs [3]. - Nvidia has also announced its own 102.4Tbps Ethernet switch, indicating a competitive landscape in high-capacity switch technology [4]. Group 3: Scalable Ethernet Solutions - Broadcom's Tomahawk 6 switches are positioned as a shortcut for rack-level architectures, supporting between 8 to 72 GPUs, with future designs expected to support up to 576 GPUs by 2027 [6]. - Ethernet technology is being utilized for both scalable and large-scale networks, with Intel and AMD also planning to implement Ethernet for their systems [7]. Group 4: Co-Packaged Optics (CPO) Technology - Broadcom has invested in co-packaged optics (CPO) technology, which integrates components typically found in pluggable transceivers into the same package as the switch ASIC, significantly reducing power consumption [9][10]. - The efficiency of Broadcom's CPO technology is reported to be over 3.5 times that of traditional pluggable devices [10]. - The third generation of CPO technology is expected to support up to 512 200Gbps optical ports, with future developments aiming for 400Gbps channels by 2028 [11]. Group 5: Multi-Chip Architecture - As Moore's Law slows, the industry is shifting towards multi-chip architectures, allowing for higher yields and optimized costs by using smaller chips [14]. - Broadcom has developed a 3.5D eXtreme Dimension System in Package (3.5D XDSiP) technology to facilitate the design of multi-chip processors, which is open for licensing to other companies [15]. - The first products based on this design are expected to enter production by 2026, although the specific applications of Broadcom's technology in AI chips may remain undisclosed [15].
小摩:AI“印钞机”全速运转 博通(AVGO.US)Q2有望交出亮眼业绩
智通财经网· 2025-06-04 07:41
智通财经APP获悉,摩根大通发表研报表示,在人工智能(AI)产品需求旺盛、非AI半导体业务趋稳以及 VMware业务持续发力的共同作用下,博通(AVGO.US)预计将公布符合或略高于市场预期的第二季度财 报,并对第三季度业绩给出乐观指引。该行维持对博通的"增持"评级,目标价为250美元。 摩根大通表示,博通的AI相关收入需求保持强劲,成为业绩增长的主要驱动力。谷歌(GOOGL.US)TPU v6 3nm ASIC芯片已开始量产爬坡,预计将为博通2025财年剩余时间及2026财年带来显著收入贡献。摩 根大通预测,该芯片项目生命周期内收入将超过150亿美元,成为全球最强定制XPU AI加速器。 在AI网络领域,博通的Tomahawk 5/Jericho 3 AI产品需求持续旺盛,下一代3nm工艺的Tomahawk 6芯片 预计于2025年下半年开始量产。综合来看,受益于谷歌TPU处理器的持续放量、Meta(META.US)/字节 跳动等其他ASIC项目的推进以及AI网络产品的稳定需求,博通2025财年AI业务收入有望达到190亿至 200亿美元,同比增长约60%。 摩根大通预计,博通第二季度营收、盈利和自由现金流 ...
102.4 Tb/s的交换机芯片,博通重磅发布
半导体行业观察· 2025-06-04 01:09
如果您希望可以时常见面,欢迎标星收藏哦~ 来源:内容 编译自 nextplatform 。 尽管随着以太网路线图上的每一次减速,更扁平的网络和更快的网络都是可能的,但网络规模仍然 保持着足够快的增长速度,以至于交换机 ASIC 制造商和交换机制造商能够通过产量来弥补这一 不足,并保持交换机业务的增长。 随着 GenAI 的爆发式增长,所有大型 AI 厂商都一致希望摆脱英伟达控制的专有 InfiniBand 技 术,将 InfiniBand 的所有功能移植到全新升级的以太网上,使其能够进一步扩展,并在更扁平的 网络中实现扩展,从而创建规模更大的 AI 集群。超级以太网联盟 (UltraEthernet Consortium) 的 宏伟目标是实现 100 万个 GPU 端点,而要实现这一目标,需要容量更大的交换机 ASIC。 如今,商用芯片市场的行业领导者博通 (Broadcom) 在以太网领域面临着来自思科系统和 Nvidia 的激烈竞争,该公司正在向市场推出其"Tomahawk 6" StrataXGS 以太网交换机 ASIC,该市场将 以 102.4 Tb/秒 ASIC 为主导,并展望 204.8 Tb/秒 ...
财报前宣布“突破性”数据中心交换机芯片,博通股价创历史新高
Hua Er Jie Jian Wen· 2025-06-04 00:47
在财报发布前48小时释放"突破性"产品消息,博通股价应声上涨近3.3%,一举突破去年12月的历史高 点,市值首次逼近1.2万亿美元。 华尔街一致看多的财报前奏 博通在财报前夕突然抛出"突破性"数据中心交换机芯片,股价创历史新高。 6月3日,当所有人都在等待博通财报时,这家芯片巨头却宣布开始出货其最新的数据中心交换机芯片 Tomahawk 6,声称其以太网交换容量达到每秒102.4太比特,是目前市场上任何以太网交换机带宽的两 倍。这款芯片专为运行AI处理器集群的数据中心需求而设计。 博通核心交换业务高级副总裁兼总经理Ram Velaga在新闻稿中强调: Tomahawk 6不仅仅是升级——它是一个突破。它标志着AI基础设施设计的转折点,将最高 带宽、功率效率和自适应路由功能结合到一个平台中。 据媒体报道,该芯片采用台积电3纳米工艺制造,相比前代Tomahawk 5的5纳米工艺实现了制程升级。 博通强调,带宽翻倍的直接好处是减少了系统所需的网络设备数量,从而降低功耗并提高可靠性。 此外,Tomahawk 6基于开放标准构建。这意味着客户不会被锁定在专有网络交换技术中,比如英伟达 推广的那些技术。这一点直击英伟达封 ...
博通CPO,重磅发布
半导体行业观察· 2025-05-17 01:54
如果您希望可以时常见面,欢迎标星收藏哦~ 来源:内容来自 Source:编译自博通 。 近 日 , 博 通 公 司 宣 布 推 出 第 三 代 单 通 道 200G ( 200G/lane ) CPO 产 品 线 , 其 共 封 装 光 模 块 (CPO)技术取得重大进展。除了200G/lane的突破外,博通还展示了其第二代100G/lane CPO产 品和生态系统的成熟度,重点突出了OSAT工艺、热设计、处理流程、光纤布线和整体良率方面的 关键改进。越来越多的行业合作伙伴已公开宣布加入,进一步凸显了博通CPO平台的成熟度,为 大规模AI部署提供AI横向扩展和纵向扩展应用。 Broadcom 在 CPO 领域的领导地位始于 2021 年,当时推出了第一代 Tomahawk 4-Humboldt 芯 片组,使整个CPO供应链能够提前学习,领先于行业。这款开创性的芯片组引入了多项关键创 新,包括高密度集成光学引擎、边缘耦合和可拆卸光纤连接器。 在此成功的基础上,第二代Tomahawk 5-Bailly(TH5-Bailly)芯片组成为业界首个量产的CPO解 决方案。在TH5-Bailly的生产过程中,博通专注于自 ...
Broadcom Announces Third-Generation Co-Packaged Optics (CPO) Technology with 200G/lane Capability
Globenewswire· 2025-05-15 13:00
Key partner milestones with CPO switches demonstrate ecosystem readiness for next-generation AI scale-up and scale-out networksPALO ALTO, Calif., May 15, 2025 (GLOBE NEWSWIRE) -- Broadcom Inc. (NASDAQ: AVGO) today announced significant advancements in its co-packaged optics (CPO) technology with the launch of its third-generation 200G per lane (200G/lane) CPO product line. In addition to the 200G/lane breakthrough, Broadcom demonstrated the maturity of its second-generation 100G/lane CPO products and ecosys ...
【招商电子】博通(AVGO.O)25Q1跟踪报告:本季新大规模XPU客户增至4家,维持AI业务长期指引
招商电子· 2025-03-09 10:49
点击招商研究小程序查看PDF报告原文 事件: 博通(NASDAQ:AVGO)发布FY2025Q1季报,第一财季营收149.16亿美元,同比+25%/环比+6%,毛 利率为79.1%,同比+3.74pcts /环比+2.2pcts。综合财报及交流会议信息,总结要点如下: 评论: 1、FY25Q1营收毛利率均超出此前预期同环比增长,库存环比有所增长。 FY25Q1营收149.16亿美元,同比+25%/环比+6%,超出此前指引(约146亿美元);毛利率为79.1%,同 比+3.74pcts /环比+2.2pcts,高于此前指引,得益于基础设施软件收入的增加和更有利的半导体收入组 合。Q1末的库存为19亿美元,较上季度增长8%,以支持未来几个季度的收入,第一季度的库存周转天 数为65天。 2、FY25Q1 AI业务营收同比+77%至41亿美元,软件受益VMwar增长明显。 1)半导体部门: 收入为82亿美元,占本季度总收入的55%,同比增长11%,毛利率约为68%,同比上 升70个基点,主要受收入结构的推动。由于在前沿AI半导体上的研发投资增加,运营费用同比增长 3%,达到8.9亿美元,导致半导体运营利润率为57%。 ...
Broadcom(AVGO) - 2025 Q1 - Earnings Call Transcript
2025-03-07 00:58
Broadcom Inc. (NASDAQ:AVGO) Q1 2025 Earnings Conference Call March 6, 2025 5:00 PM ET Company Participants Ji Yoo - Head, IR Hock Tan - President & CEO Kirsten Spears - CFO Conference Call Participants Ben Reitzes - Melius Research Harlan Sur - JPMorgan William Stein - Truist Securities Ross Seymore - Deutsche Bank Stacy Rasgon - Bernstein Vivek Arya - Bank of America Harsh Kumar - Piper Sandler Timothy Arcuri - UBS CJ Muse - Cantor Fitzgerald Christopher Rolland - Susquehanna Vijay Rakesh - Mizuho Securiti ...