Nvidia (NVDA) printed an all-time high of $191.05 on Thursday as investors continued to cheer its $100 billion partnership with OpenAI, cementing its leadership in artificial intelligence (AI) hardware. Shares of the chipmaker have pushed higher in recent weeks also because the Federal Reserve lowered interest rates by 25 basis points on Sept. 17, triggering a broader rally in the high-flying tech stock. More News from Barchart Including today’s gains, Nvidia stock is up roughly 120% versus its year-to- ...

Core Viewpoint - The article discusses the advancements in GPU communication technologies, particularly focusing on GPUDirect Storage, GPUDirect P2P, NVLink, NVSwitch, and GPUDirect RDMA, which enhance data transfer efficiency and reduce bottlenecks in high-performance computing environments [27]. Group 1: GPU and Storage Communication - The data flow from storage systems to GPU memory involves two data copies: from NVMe SSD to system memory and then from system memory to GPU memory, which introduces redundancy [6]. - GPUDirect Storage allows direct access from storage to GPU memory, significantly improving data loading efficiency by reducing unnecessary system copies [7]. Group 2: GPU to GPU Communication - Traditional GPU communication involves multiple data copies through system memory, which can be inefficient [10]. - GPUDirect P2P enables direct data transfer between GPUs, bypassing the CPU and reducing data copy actions by half [12]. Group 3: NVLink and NVSwitch - NVLink provides high bandwidth for data transfer between GPUs, achieving up to 600GB/s for NVIDIA A100 Tensor Core GPUs, which is significantly higher than traditional PCIe [15]. - NVSwitch facilitates full interconnectivity among multiple GPUs, supporting high bandwidth and scalability for large GPU systems [20]. Group 4: Cross-Machine Communication - Traditional cross-machine communication requires multiple steps involving system memory, which can be inefficient [22][24]. - GPUDirect RDMA simplifies this process, allowing direct access to GPU memory from peripheral PCIe devices, thus enhancing communication efficiency [25]. Group 5: Summary of Technologies - The combination of GPUDirect technologies, including P2P and RDMA, supports efficient communication within single nodes and across multiple nodes, essential for AI training and high-performance computing [28].

The Nvidia logo is displayed on a building at Nvidia headquarters on Aug. 27, 2025 in Santa Clara, California.Despite macroeconomic uncertainties, several companies are well-positioned to deliver strong returns to investors from rapid technological advancements and artificial intelligence (AI) adoption. To pick attractive stocks with strong prospects, investors can track top Wall Street analysts, whose recommendations are based on in-depth research and analysis of a company's financials and growth drivers.H ...

2025年9月24日，中国信息通信研究院华东分院与行业领先的AI网络全栈式互联产品及解决 方案提供商——奇异摩尔联合举办的"聚力向芯 算涌无界 Networking for AI"生态沙龙活动 在上海浦东成功举办。 活动围绕"国产AI算力产业链实现闭环"这一主题，邀请AI算力产业链上下游企业及技术专家，包 括中国移动、财跃星辰、腾讯云、沐曦、新华三、中科创星、无问芯穹、后摩智能、中昊芯英等行 业头部及明星企业。内容覆盖从算力芯片到算法模型再到算力服务的全链条环节，展现了国产AI 从技术突破到生态构建的整体进展，也标志着中国在科技自主化道路上的坚实脚步。 "聚力向芯 算涌无界 Networking for AI" 生态沙龙活动 在开场致辞环节，中国信通院华东分院总工程师陈俊琰表示，"在国家战略指引和地方政策推动 下，上海智能算力产业已形成'基础夯实、创新活跃、场景丰富'的良好态势。面向上海要实现智能 算力产业高质量发展，他提出了三点建议：一是强化技术创新协同，突破核心瓶颈。二是完善算力 调度体系，提升配置效率。三是深化场景融合应用，赋能实体经济。中国信通院华东分院将继续发 挥'国家高端专业智库、产业创新发展平台 ...

分组1: Nvidia and Intel Deal - Nvidia has signed a deal to take a $5 billion equity stake in Intel, co-developing custom products for data centers and personal computers [1][2] - Nvidia holds a 90% market share in GPUs but is not the leader in CPUs, which are dominated by Intel and AMD [2] - The deal aims to enhance communication between Nvidia's GPUs and Intel's CPUs through NVLink Fusion, allowing for faster data transfer [2] 分组2: Market Implications - The partnership is seen as a strategic move by Nvidia and Intel to counter AMD's market share in CPUs and GPUs [5] - Nvidia's investment represents a small fraction (0.1%) of its market cap, indicating that while significant, it may not drastically alter Nvidia's overall strategy [4] - There are potential regulatory challenges regarding the partnership, as it could be viewed as anti-competitive [4] 分组3: Quarterly Earnings Discussion - The discussion around quarterly earnings reports highlights their importance for early-stage companies to provide regular performance snapshots [7][8] - Some experts argue for a shift to semi-annual reporting to reduce compliance costs, although this may not significantly change short-term business focus [7][8] - Concerns exist that less frequent reporting could allow issues to fester longer, potentially harming investors [9] 分组4: Stocks on the Radar - The Trade Desk is being monitored due to its recent struggles despite a strong historical performance, with management addressing user concerns about its new AI platform [12] - General Motors is highlighted for its strong position in the EV market and aggressive stock buybacks, trading at a low PE ratio [14] - Celestica is noted for its role in the AI infrastructure boom, benefiting from increased demand for assembly and manufacturing services [15][16]

Group 1 - Nvidia's recent $5 billion investment in Intel indicates that Nvidia is entering AMD's business, particularly in the area of system on chips [1][2] - The collaboration between Nvidia and Intel aims to integrate GPUs and CPUs, enhancing capabilities for gaming desktops and AI laptop processors [2] - AMD is expected to improve its market share in AI applications by 2027, potentially matching the market leader, Nvidia, as hyperscale customers may opt for dual-sourcing high-end chips [3] Group 2 - Macquarie Core Equity Fund acknowledges AMD's potential in the semiconductor market, particularly in CPUs and GPUs for gaming, data centers, and AI [3] - Despite recognizing AMD's growth potential, the fund believes that other AI stocks may offer higher returns with limited downside risk [3]

如果您希望可以时常见面，欢迎标星收藏哦~ 来源：内容来自 technews 。 人工智能（AI）运算能力的军备竞赛持续升温，全球基础设施投资正迎来史上最大规模的浪潮。 然而，AI 产业在处理器之间数据移动方面，正遭遇「一堵根本性的高墙」。因为现有采用铜线的 互连技术，已无法有效扩展，进一步服务新世代AI 所需的数百万个处理器。在此背景下，光子芯 片（Photonic Chips）新创公司Celestial AIAI 和OpenLight 正迅速崛起，提供更快速、功耗更低 的解决方案，以满足亚马逊、微软和Google 等超大规模（hyperscaler）客户的需求。 Celestial AI有台积电与三星投资， 锁定伺服器纵向扩展市场 为应对市场对下一代AI 基础设施飙升的需求，光子芯片新创公司获得了强大的资金支持并深化了 与代工伙伴的合作。其中，Celestial AI 在2025 年8 月份获得了来自台积电相关的VentureTech Alliance 和三星Catalyst Fund 的2.55 亿美元投资，使其总募资金额达到5.2 亿美元。 Celestial AI计划利用这笔新资金加强其供应链，并深化 ...

We recently published Top 10 Stocks to Watch After Federal Reserve’s Rate Cut. Advanced Micro Devices, Inc. (NASDAQ:AMD) is one of the stocks analysts were recently talking about. Robert Schein, CIO of Blanke Schein Wealth Management, said in a program on CNBC in late July that AMD will give Nvidia a “run for its money.” Here is how the analyst explained his bullish thesis for the chip company. “Valuation is a terrible tool for trading. we look at valuations. put that aside. We want to look at so what is ...

Summary of Conference Call on NVIDIA and Intel Partnership Industry Overview - **Industry**: Greater China Technology Semiconductors - **Region**: Asia Pacific Key Company Insights - **NVIDIA** plans to invest **US$5 billion** in **Intel** common stock, indicating a strong partnership between the two companies [2][6] - **Intel** will continue its CPU foundry outsourcing to **TSMC**, which may negatively impact ARM-based AI PCs and AMD's laptop CPUs initially, with potential future effects on **ASMedia** in desktop PC chipsets [1][6] Core Partnership Developments 1. **Product Development**: The partnership aims to jointly develop multiple generations of custom data center and PC products, focusing on integrating NVIDIA and Intel architectures using **NVIDIA NVLink** [6][10] 2. **Market Impact**: - Intel will produce NVIDIA-custom x86 CPUs for data centers, which will be integrated into NVIDIA's AI infrastructure platforms [6][10] - Intel will also offer x86 SoCs that integrate NVIDIA RTX GPU chiplets, expanding their market reach [6][10] 3. **Market Addressability**: NVIDIA highlighted an addressable market of **150 million** notebook units, indicating significant potential for the new x86 SoC that combines Intel CPU and NVIDIA GPU [6][10] Implications for Competitors - **MediaTek**: Slightly negative outlook for MediaTek's WoA in the gaming/AI PC market due to the new x86 SoC fusing Intel CPU and NVIDIA GPU [6][10] - **ASMedia**: The direct impact on AMD's desktop CPU share is expected to be limited, but overall sentiment may be negative for ASMedia due to potential future collaborations between Intel and NVIDIA [6][10] Strategic Insights - **TSMC's Role**: The partnership does not change NVIDIA's ARM roadmap, but emphasizes TSMC's strong foundry capabilities and Intel's advanced packaging capabilities, which could enhance the integration of NVIDIA GPU chiplets with Intel CPUs [6][10] - **Market Expansion**: The partnership is expected to expand the total addressable market (TAM) for both companies, with Intel gaining exposure to NVLink-based AI infrastructure and premium laptop CPUs, while NVIDIA gains access to the laptop graphics segment and x86 rack-scale solutions [6][10] Risks and Considerations - **Market Dynamics**: Potential risks include increased competition from AMD, supply chain diversification, and the impact of US-China tensions on semiconductor supply chains [15][19] - **Financial Outlook**: Intel's 2026 EPS is projected at **~29x** with a focus on regaining market share in desktop and server segments following CPU shortages [17][19] Conclusion The partnership between NVIDIA and Intel represents a significant strategic move in the semiconductor industry, with implications for various players in the market. The collaboration aims to leverage both companies' strengths to create innovative products while navigating potential competitive and market risks.

Group 1 - Nvidia announced a $5 billion investment in Intel to collaborate on new data center and PC products, combining Nvidia's computing stack with Intel's x86 CPUs [1][4] - The partnership is significant for Intel as it provides a much-needed endorsement during its turnaround, while Nvidia aims to expand its platform beyond GPUs [2][5] - Nvidia will purchase Intel stock at $23.28 per share, and Intel will design custom x86 CPUs for Nvidia's AI platforms [4][5] Group 2 - The collaboration has immediate implications for data centers, as Intel will build custom x86 CPUs for Nvidia's AI platforms, potentially improving bandwidth and reducing latency [6] - Intel's business faces ongoing pressure, with declining margins and lost market share, making this partnership crucial for restoring credibility [7] - The alliance could reshape competition in both data centers and PCs, as Nvidia continues to grow while Intel is restructuring [8]