Report Industry Investment Rating No relevant information provided. Core Views of the Report - AI generalization has entered a large-scale application period, leading to an exponential increase in computing power demand across various industries. The US is facing a power crisis, and China is likely to experience a power shortage in 1 - 2 years. China will win the Sino-US AI game, and the semiconductor equipment and energy storage industries will be highly prosperous. Silver is a commodity that benefits from both endless computing power and endless electricity. The stock market style is shifting, and the CSI 300 Index is well-balanced. Gold will soar due to the Fed's policy shift, and the demand for commodities will decline. The RMB exchange rate is expected to strengthen [1][2][3]. Summary by Relevant Catalogs Economic Cycles - **Kitchin Cycle**: China's current Kitchin cycle is expected to reach its peak in Q4 2025, while the US will reach its peak in Q1 2026 [6]. - **Juglar Cycle**: China's Juglar cycle is currently in an upward phase and is expected to peak in early 2027 [7]. - **Kuznets Cycle**: China's current Kuznets cycle is expected to bottom out around 2030 [8]. - **Kondratieff Cycle**: The current Kondratieff depression started in 2020 due to the COVID-19 impact and is expected to end around 2030, followed by a 10-year recovery period. China is the center of this technological innovation cycle, with AI being the most significant innovation [9]. AI Computing Power Demand - The implementation of the "AI +" action indicates China's full transition to an AI economy. The demand for computing power is growing exponentially, with token usage expected to increase by about 30 times in a year, 900 times in two years, and 27,000 times in three years. Tech giants are making substantial investments in computing power infrastructure, and global AI data center and chip investment is expected to reach $2.9 trillion by 2028 [10][11][14]. Power Demand and Crisis - The exponential growth in computing power demand leads to an exponential increase in power demand, intensifying the Sino-US power competition. The US is facing a power crisis as its aging power grid struggles to meet the growing demand. China is likely to experience a power shortage in 1 - 2 years despite its efforts to expand new energy power generation [15][16][23]. Semiconductor Equipment Industry - Huawei's AI chip computing power supernodes lead the industry, indicating China's full autonomy in the chip industry chain. The surging demand for computing power will drive the semiconductor equipment industry to remain highly prosperous, and related ETFs are expected to perform well [24][25][28]. Energy Storage and Photovoltaic Industries - China's strong power construction mobilization ability ensures its victory in the Sino-US AI game. The energy storage industry is highly prosperous due to the increasing power demand, and the photovoltaic industry is expected to recover in 2026 [33][36]. Silver - Silver benefits from both endless computing power and endless electricity. The demand for silver is expected to surge, widening the supply-demand gap and leading to a shortage of physical silver [37]. Stock Market - The stock market style is shifting, and the CSI 300 Index is well-balanced with limited downside and significant upside potential. The CSI 500 and CSI 300 indices have reached new highs, and the Sci-Tech Innovation 50 Index and ChiNext Index are expected to continue rising. A large amount of domestic and foreign capital is flowing into the A-share market [44][46][47]. Gold - The Fed's policy shift towards full employment and tolerance of inflation, combined with the potential for negative real interest rates in the US, will drive gold prices higher [48][49]. Commodities - Due to the anti-involution in the mid - and downstream industries, the demand for upstream resource commodities is weakening. However, copper prices may strengthen temporarily, and lithium carbonate may enter a new bull market in H2 2026 [50][51][52]. Bonds - As the stock market rises, bond funds are facing large-scale redemptions, and the bond market is becoming a source of funds for the stock market. The real interest rate is trending towards negative values [52]. Foreign Exchange - China is expected to achieve a double surplus in trade and capital, and the offshore RMB exchange rate is likely to enter the 6 - level by the end of the year [53][55][56]. October Outlook for Major Asset Classes - **Equity Assets**: The stock market's wealth effect is attracting savings from the household sector, and overseas funds are flowing into Chinese assets. The semiconductor equipment index and battery index are expected to remain highly prosperous [3][56]. - **Gold and Silver**: Gold will soar due to the Fed's policy shift, and silver may face severe supply shortages [3][56]. - **Commodities**: The demand for upstream resource commodities is weakening [3][56]. - **Bond Assets**: The bond market is becoming a source of funds for the stock market as the real interest rate turns negative [3][56]. - **Foreign Exchange Assets**: The offshore RMB exchange rate is expected to strengthen and enter the 6 - level by the end of the year [3][56].

Group 1 - Company provides cooling systems for packaging lithography machines for Shanghai Micro Electronics Equipment Group, with a revenue increase from 625.83% to 756.71% [2] - Company offers lithography objectives, contributing to the optical industry [2] - Major US storage leaders have reached historical highs; Taiwanese manufacturers have fully booked DDR4 production capacity for Q4 [2] - Company’s enterprise-level storage is primarily used in servers, data centers, and cloud computing [2] - Huawei plans to launch the world's strongest super node in Q4, with a computing power scale of 8192 cards [2] - Company is a leading player in the global fiber optic cable industry, with advanced hollow core technology [2] - Company’s Micro TEC products have been validated by leading enterprises and are now in mass production [2] - Company is expanding its business in the energy storage sector through precision box systems [2] Group 2 - Company’s controlling shareholder has changed to Yang Gongyi Fan [3] - Company develops chips using the CK802 32-bit core architecture from Pingtouge [3] - Company’s photoresist substrate products are currently in the customer engagement phase [3] - Company has signed a strategic cooperation agreement with Suzhou Nengda to jointly develop flexible electronic skin [3] - Company focuses on the development of intelligent cleaning robots based on a skateboard chassis [3] - Company invests in smart elderly care robot projects, integrating AI with home care [3] Group 3 - Company is a hardware supplier for Leju Robotics [4] - Company participates in the information technology project for the C919 large aircraft [4] - Company is a leading cinema investment and operation brand in China [4] - Company has signed a strategic cooperation framework with Huawei to collaborate in network, cloud computing, and big data [4] Group 4 - Company is a major player in the historical cultural scenic area development, managing several national 5A scenic spots [5] - Company is the second-largest shareholder of Tianke He Da, which is a global leader in SiC chip manufacturing [5] - Company’s main products include silicon carbide crystal growth furnaces [5] - Company has seen price increases in express delivery services across multiple regions [5] Group 5 - Company has been awarded a project for intelligent cockpit domain controllers, with a total amount of approximately 5.6 billion [6] - Company is a leading construction firm in Xinjiang with comprehensive qualifications [6] - Company’s controlling shareholder has changed to Yang Gongyi Fan [6] - Company is deeply involved in AI computing server connections and assembly for Apple hardware [6] Group 6 - Company has a long-standing partnership with Huawei on smart water projects [7] - Company’s subsidiary has mining rights for a gold mine, with an expanded mining scale approved [7] - Company produces silver ion antibacterial fabrics using SILVER+ technology [7]

2025年中国国际信息通信展（业内也称为"北展"）开展在即，作为国内通信行业的风向标，如何在当前 与AI深度融合的浪潮下顺利转型，成为本届展前全行业最为关注的话题。 记者在现场了解到，目前华为5G-A相关的套餐与终端生态已进入规模验证阶段，部分差速化套餐用户 增长明显。 而在家庭端，华为则通过FTTR、50G PON及Wi-Fi7组合，把家庭打造为"AI to H"的本地化承载体，从 而为语音/视频交互、本地轻量化推理提供低时延支撑。对此，梁策表示，F5G-A 正在将家庭从"居住空 间"升级为"智能体中心"，后续还将实现生鲜直购等生活服务闭环。未来，运营商将通过不同档位的差 速化套餐和家庭盒子等终端形态，尝试把体验升级转化为付费意愿。 最被业界关注的，则是华为关于"超节点"与省市级算力池的布局。据华为智算商业发展部部长王法透 露，384超节点更适配MoE（多专家架构）及低时延大规模推理需求，目前华为及运营商已建立超节 点，且运营商省市机房已成为主流算力部署形式，入驻式大机房的比例显著下降。王法强调，算力的地 理分布、就近服务的合规与效率属性，使运营商天然具备成为边缘智算提供方的条件。 以算助联、以联促算 对此 ...

Macro Strategy - The core viewpoint indicates that the recent cooling of U.S. employment data makes a rate cut in September almost certain, with expectations of a 25bps cut and potential for 1-2 additional cuts throughout the year [1][14][18] - The U.S. non-farm payrolls for August showed an increase of only 22,000, significantly below the expected 75,000, indicating a weakening labor market [1][18] - The unemployment rate rose to 4.324%, slightly above expectations, highlighting a trend of weakening labor demand [1][18] Fixed Income - In the week of September 1-5, 19 green bonds were issued in the interbank and exchange markets, totaling approximately 8.767 billion yuan, an increase of 1.651 billion yuan from the previous week [2] - The secondary market saw a total trading volume of green bonds amounting to 48.2 billion yuan, a decrease of 4 billion yuan from the previous week [2] Industry Analysis - The domestic aluminum electrolytic capacitor industry is experiencing stable growth, with a focus on supercapacitors expected to drive a second growth curve for Jianghai Co., Ltd. [10] - Jianghai Co., Ltd. is positioned as a leader in the supercapacitor market, with a compound annual growth rate of 23.9% in the domestic market from 2017 to 2023 [10] - The report highlights the potential for significant growth in the supercapacitor business, particularly in AI computing applications, as the company collaborates with multiple AI clients [10] Company Recommendations - Yapu Co., Ltd. is expected to see revenue growth from 9.123 billion yuan in 2025 to 10.961 billion yuan in 2027, with a corresponding net profit increase from 575 million yuan to 756 million yuan [9] - Jianghai Co., Ltd. is projected to achieve revenues of 4.565 billion yuan in 2025, growing to 5.770 billion yuan by 2027, with net profits expected to rise from 791 million yuan to 1.025 billion yuan [13] - China Pacific Insurance is maintaining a positive outlook with projected net profits of 51.6 billion yuan in 2025, increasing to 55.3 billion yuan by 2027 [12]

Macro Strategy - The recent cooling of US employment data makes a rate cut in September almost certain, with expectations of a 25bps cut and potential for 1-2 additional cuts throughout the year [1][12][16] - Gold prices surged, breaking the $3600 per ounce mark, driven by concerns over Eurozone fiscal stability and US rate cut expectations [1][12] - Upcoming US non-farm payroll data and inflation metrics (PPI, CPI) will significantly influence the tone of the September FOMC meeting [1][12][16] Employment Data Analysis - In August, the US added only 22,000 non-farm jobs, significantly below the expected 75,000, with previous months' data also revised downwards [2][16] - The unemployment rate rose to 4.324%, slightly above expectations, indicating a weakening labor demand [2][16] - The labor market is characterized by a "bizarre balance" of weak supply and demand, leading to a more pronounced decline in job additions compared to the rise in unemployment [2][16] Fixed Income Market Insights - The issuance of green bonds increased to approximately 8.767 billion yuan, reflecting a growing interest in sustainable finance [3] - The secondary market for green bonds saw a decrease in trading volume, totaling 48.2 billion yuan, indicating a potential cooling in investor interest [3] - The bond market is experiencing a "stock-bond seesaw" effect, with limited downward movement in bond yields despite significant stock market corrections [4][5] Industry Analysis: Semiconductor Sector - The domestic computing power sector is showing signs of recovery, with increased activity in GPU orders from major internet companies [10] - Companies like Cambrian and Haiguang are actively pursuing capital operations and acquisitions, indicating a bullish sentiment in the semiconductor space [10] - The release of new high-performance chips is expected to drive further growth in the domestic computing power market, with significant implications for internet and ODM companies [10] Company-Specific Insights: NIO Inc. - NIO's new vehicle performance is strong, but the company has revised its net profit forecasts for 2025-2026 downwards due to increased industry competition [11] - Despite the downward revision, NIO maintains a "buy" rating based on its comprehensive technology layout and marketing strategies [11]

Core Insights - The rapid development of large models driven by the AI wave has created stringent demands for computing power, leading to the emergence of the "SuperPod" as a key solution in the industry [1][2] - The transition from traditional computing architectures to SuperPod technology signifies a shift towards high-performance, low-cost, and energy-efficient AI training solutions [1][2] Industry Trends - The SuperPod, proposed by NVIDIA, represents the optimal solution for Scale Up architecture, integrating GPU resources to create a low-latency, high-bandwidth computing entity [2] - The traditional air-cooled AI servers are reaching their power density limits, prompting the adoption of advanced cooling technologies like liquid cooling in SuperPod designs [2][5] - The market outlook for SuperPods is optimistic, with many domestic and international server manufacturers adopting this next-generation solution [2][4] Technological Developments - Current mainstream SuperPod solutions include private protocol schemes (e.g., NVIDIA, Trainium, Huawei) and open organization schemes, with copper connections becoming increasingly prevalent for internal communications [3][4] - The ETH-X open SuperPod project, led by the Open Data Center Committee, exemplifies the integration of Scale Up and Scale Out networking strategies [4] Company Initiatives - Chinese tech companies are actively investing in the SuperPod space, with Huawei showcasing its Ascend 384 SuperPod, which features the largest scale of 384-card high-speed bus interconnection [5] - Other companies like Xizhi Technology and Muxi have introduced innovative solutions, such as distributed optical interconnects and liquid-cooled GPU modules, enhancing the SuperPod technology landscape [5][6] - Moore Threads has established a comprehensive AI computing product line, aiming to create a new generation of AI training infrastructure, referred to as a "super factory" for advanced model production [6]

Core Insights - The recent WAIC highlighted the growing interest in supernodes, with companies like Huawei, ZTE, and H3C showcasing their advancements in this technology [3][4][5] - Supernodes are essential for managing large-scale AI models, enabling efficient resource utilization and high-performance computing [3][4][5] - The shift from traditional AI servers to supernode architectures is driven by the increasing complexity and size of AI models, which now reach trillions of parameters [4][5][9] Group 1: Supernode Technology - Supernodes integrate computing resources to create low-latency, high-bandwidth computing entities, enhancing the efficiency of AI model training and inference [3][4] - The technology allows for performance improvements even when individual chip manufacturing processes are limited, making it a crucial development in the industry [4][9] - Companies are exploring both horizontal (scale out) and vertical (scale up) expansion strategies to optimize supernode performance [5][9] Group 2: Market Dynamics - Domestic AI chip manufacturers are increasing their market share in AI servers, with the proportion of externally sourced chips expected to drop from 63% to 49% this year [10] - Companies like墨芯人工智能 are adopting strategies that focus on specific AI applications, such as inference optimization, to compete with established players like NVIDIA [10][11] - The competitive landscape is shifting, with firms like云天励飞 and后摩智能 targeting niche markets in edge computing and AI inference, avoiding direct competition with larger chip manufacturers [11][12][13] Group 3: Technological Innovations - The introduction of optical interconnects in supernode technology is a significant advancement, providing high bandwidth and low latency for AI workloads [6][9] - Companies are developing solutions that leverage optical communication to enhance the performance of AI chip clusters, addressing the limitations of traditional electrical interconnects [6][9] - The focus on sparse computing techniques allows for lower manufacturing process requirements, enabling more efficient AI model computations [11][12]

Investment Rating - The report maintains a positive outlook on the computer industry, particularly focusing on the developments in stablecoins and supernodes [5][6]. Core Insights - The report emphasizes that stablecoins are transitioning from speculative concepts to mainstream payment solutions, supported by regulatory developments and strategic initiatives in regions like Shanghai and Hong Kong [7][8]. - The supernode architecture is highlighted as a significant advancement in computational power, with products like Huawei's CloudMatrix 384 demonstrating superior performance compared to traditional single-card systems [5][33]. Summary by Sections Stablecoins - The Shanghai Municipal State-owned Assets Supervision and Administration Commission recognizes the role of stablecoins in enhancing the financial system and encourages the exploration of related technologies [7][8]. - The report outlines that stablecoins are not merely a short-term trend but represent a long-term investment theme, with a shift towards legislative compliance and mainstream payment applications [6][9]. - Key companies involved in stablecoin technology and infrastructure include 恒生电子 (Hengsheng Electronics), 金证股份 (Jinzhen Co.), and others [5][6]. Supernodes - The report discusses the transition from single-card systems to supernode architectures, which integrate multiple GPUs for enhanced computational efficiency [33][34]. - Huawei's CloudMatrix 384 is presented as a benchmark for domestic supernode solutions, achieving 1.7 times the computational power and 3.6 times the memory capacity compared to Nvidia's GB200 NVL72 [55][56]. - Companies positioned to benefit from the supernode trend include 海光信息 (Haiguang Information), 浪潮信息 (Inspur Information), and 神州数码 (Digital China) [5][33].

Core Viewpoint - The article discusses the emergence of supernodes in high-performance computing, emphasizing their role in enhancing the efficiency of large-scale model training and inference through optical technology [1][2][21]. Group 1: Supernode Architecture and Performance - Supernodes provide a new solution for large-scale model training and inference, significantly improving efficiency by optimizing resource allocation and data transmission [1]. - The architecture of supernodes can be categorized into single-layer and two-layer designs, with single-layer architecture being the ultimate goal due to its lower latency and higher reliability [4][6]. - The demand for GPU power has surged with the exponential growth of model sizes, necessitating thousands of GPUs to work in tandem, which supernodes can facilitate [1][2]. Group 2: Challenges in Domestic Ecosystem - Domestic GPUs face significant performance gaps compared to international counterparts, requiring hundreds of domestic GPUs to match the power of a few high-end international GPUs [6][8]. - The implementation of supernodes in the domestic market is hindered by limitations in manufacturing processes, such as the 7nm technology [6]. Group 3: Development Paths for Supernodes - Two main development paths are proposed: increasing the power capacity of individual cabinets to accommodate more GPUs or increasing the number of cabinets while ensuring efficient interconnection [8][10]. - Optical interconnect technology is crucial for multi-cabinet scenarios, offering significant advantages over traditional copper cables in terms of transmission distance and flexibility [10][12]. Group 4: Optical Technology Advancements - The transition to higher integration optical products, such as Co-Packaged Optics (CPO), enhances system performance by reducing complexity and improving reliability [14][16]. - CPO technology can save 1/3 to 2/3 of power consumption, which is significant even though communication power is a smaller fraction of total GPU power [16][17]. Group 5: Reliability and Flexibility - The use of distributed optical switching technology enhances the flexibility and reliability of supernodes, allowing for dynamic topology adjustments in case of node failures [18][19]. - Optical interconnect technology simplifies the supply chain, making it more controllable compared to advanced process-dependent components [19][21]. Group 6: Future Outlook - With advancements in domestic GPU performance and the maturation of optical interconnect technology, the supernode ecosystem is expected to achieve significant breakthroughs, supporting the rapid development of artificial intelligence [21].