Core Viewpoint - The integration of cloud computing and intelligent computing is reshaping the underlying logic of the digital economy, with a focus on the need for on-demand access to computing resources driven by advancements in artificial intelligence technology [1][2]. Group 1: Policy Framework and Development - The policy framework for the computing power industry is gradually improving, with numerous institutional documents being released to solidify the foundation for high-quality and large-scale development [2]. - The "Action Plan for High-Quality Development of Computing Power Infrastructure" aims to establish a unified system for computing resource identification and authentication, while enhancing security capabilities [2][12]. - The Ministry of Industry and Information Technology issued the "Action Plan for Computing Power Interconnection," emphasizing the need for a secure computing power internet and integrated security capabilities across cloud, network, edge, and terminal [2][12]. Group 2: Security Challenges in Computing Power Internet - The computing power internet is a new network technology system designed for compute-intensive tasks and dynamic scheduling, requiring a unified identification mechanism and standardized protocol interfaces [3]. - The current security system of the computing power internet is still immature, facing risks such as resource hijacking, denial of service, network channel attacks, and model poisoning [3][4]. - The complexity of the computing power interconnection system necessitates urgent and systematic risk prevention and governance measures [3]. Group 3: Infrastructure and Technical Risks - The security risks at the computing power infrastructure layer stem from technological heterogeneity and weak interconnection mechanisms, leading to compatibility issues and immature security systems [4][5]. - The complexity of computing resource architecture, including heterogeneous chips, complicates compatibility and increases vulnerability to attacks [5]. - The core interconnection system lacks a unified identification system, making it difficult to prevent risks such as forgery and tampering [5][6]. Group 4: Resource Interconnection and Management Challenges - Security challenges at the resource interconnection layer include difficulties in identity management and insufficient trust in scheduling [6][8]. - The complexity of managing identities and permissions on large resource scheduling platforms increases the risk of unauthorized access and cross-domain attacks [6][8]. Group 5: Application Layer Stability and Data Security - Ensuring the stability and low-latency response of computing power application services is challenging, particularly during surges in demand or node failures [10]. - The division of data security responsibilities among multiple stakeholders, including users, service providers, and regulatory bodies, complicates accountability and increases risks of data breaches [10][12]. Group 6: Building a Secure and Trustworthy System - The construction of a trustworthy computing power internet system is essential for addressing non-traditional security challenges and promoting high-quality digital economic development [12]. - Measures include developing a domestic computing card ecosystem, establishing unified identification systems, and enhancing network transmission technologies [12].

Core Insights - The National Cybersecurity Base (Wuhan) Computing Power Center has officially connected to the National Supercomputing Internet, marking a significant step in integrating regional computing resources into a national network [1][3] - The "East Data West Computing" strategy emphasizes the importance of computing power as a fundamental resource, akin to water and electricity [1] - The center will provide substantial support for technological innovation and industrial upgrades in Wuhan, particularly in advanced research fields and traditional industries [3][5] Group 1: Technological Innovation - The center will support major research projects in fields such as aerospace, biomedicine, and meteorology by providing powerful supercomputing resources to local universities and research teams [3] - It aims to accelerate original innovation and the transformation of research outcomes through enhanced computational capabilities [3] Group 2: Industrial Transformation - The center will offer accessible high-performance computing support to Wuhan's traditional manufacturing, health, and new display industries [3] - It will facilitate complex simulation design, AI model training, and big data analysis, promoting the integration of "AI+" and "Data+" with traditional manufacturing [3] Group 3: Network Security Enhancement - The center is designed with special network security features and capabilities, leveraging large-scale parallel computing for deep analysis of network traffic and session logs [5] - It has launched multiple cybersecurity applications for internet content monitoring and data analysis, and will support cybersecurity competitions with an efficient environment [5] Group 4: Infrastructure and Technology - The center utilizes over a thousand domestic servers built with domestic X86 processors and GPGPU acceleration cards, ensuring self-control and security while mitigating foreign chip risks [3] - It employs immersion liquid cooling technology, enhancing heat exchange efficiency and allowing for a 15%-20% increase in CPU frequency while reducing energy costs [4]

Core Insights - The Ministry of Industry and Information Technology (MIIT) is optimizing the layout of computing power infrastructure, which is expected to accelerate the construction of intelligent computing centers and drive the growth of domestic computing power demand [1][2]. Group 1: Policy and Infrastructure Development - As of June 2025, the number of standard racks in use for computing power centers in China reached 10.85 million, with intelligent computing power scaling up to 788 EFLOPS [1]. - The MIIT plans to improve the policy system for computing power layout and guide local governments in rationally establishing intelligent computing facilities [1][2]. - The official connection of the Chinese computing power platform includes ten provinces and cities, such as Shanxi, Zhejiang, and Shanghai, which is expected to enhance the overall layout of computing power centers across the country [2]. Group 2: Market Implications - The integration of the national supercomputing internet and the connection of seven computing power centers is anticipated to promote the development of the domestic AI industry and increase demand for domestic computing power [2]. - The acceleration of local intelligent computing center investments is seen as a catalyst for growth in the computing power sector [3].

Core Viewpoint - The Chinese computing power industry is transitioning from scale expansion to high-quality development, with significant advancements in infrastructure, technological innovation, and ecological construction [1]. Group 1: Industry Development - As of June 2023, China has 10.85 million standard racks in use for computing power centers and an intelligent computing power scale of 788 EFLOPS [1]. - The total storage capacity in China has exceeded 1680 EB, indicating continuous growth in storage capabilities [1]. - The "East Data West Computing" initiative plans to construct over 250 optical cable trunk lines to enhance data transmission [1]. Group 2: Technological Breakthroughs - The 2025 China Computing Power Conference highlighted 10 major technological breakthroughs, including the "China Mobile 'Jiuzhou' Computing Power Optical Network" and "Shanxi Qineng Integrated Platform" [2]. - Key core technologies in chips and software have made significant progress, contributing to the global leadership of China's computing power sector [2]. Group 3: Infrastructure and Connectivity - Ten provinces and cities, including Shanxi and Shanghai, announced their connection to the national computing power platform, forming a new development pattern of "overall overview and regional collaboration" [3]. - The number of backbone node computing centers connected to the national supercomputing internet has surpassed 30, with over 200 members in the consortium [3]. Group 4: Environmental Sustainability - The average Power Usage Effectiveness (PUE) of Chinese computing centers has decreased to 1.42, with 246 national green data centers established to support industrial upgrades and achieve carbon neutrality goals [3]. - The National Advanced Computing Taiyuan Center has implemented a new generation of immersion phase change liquid cooling technology, achieving a PUE value of 1.04 [3]. Group 5: Future Outlook - The future development of the Chinese computing power industry will focus on enhancing quality and speed, deepening the construction of an integrated computing power network, and optimizing resource coordination [4]. - Policies will be promoted to improve the layout of computing power infrastructure and accelerate breakthroughs in key core technologies, particularly in education, healthcare, and energy sectors [4].

Core Insights - The Chinese computing power industry is transitioning from scale expansion to high-quality development, as highlighted by the Ministry of Industry and Information Technology [1] - The 2025 Computing Power Development Report indicates that China's computing power infrastructure ranks among the top globally, with significant advancements in computing, storage, and transmission capabilities [1][2] Infrastructure and Capacity - As of June 2023, China has 10.85 million standard racks in use for computing power centers and an intelligent computing power scale of 788 EFLOPS [1] - The total storage capacity has exceeded 1680 EB, reflecting continuous growth in storage capabilities [1] - Over 250 "East Data West Computing" optical cable trunk lines are planned for construction [1] Technological Innovations - The forum announced 10 major breakthroughs in computing power technology, including significant contributions from China Mobile, Shanxi Qineng Technology, and Huawei [2] - The report emphasizes that key core technologies in chips and software have made breakthrough progress [2] Integration and Collaboration - The construction of an integrated computing power network is progressing, with 10 provinces connecting to the national computing power platform [3] - The number of supercomputing centers connected to the national supercomputing internet has surpassed 30, with over 200 members in the consortium [3] Environmental Sustainability - The average Power Usage Effectiveness (PUE) of Chinese computing centers has decreased to 1.42, with 246 national green data centers established [3] - The use of advanced liquid cooling technology at the National Advanced Computing Taiyuan Center has achieved a PUE value of 1.04, contributing to energy efficiency [3] Future Development - The industry aims to enhance quality and speed, deepen the construction of an integrated computing power network, and optimize resource coordination and dynamic capabilities [4] - Policies will be developed to improve the layout of computing power infrastructure and promote the application of computing power in key sectors such as education, healthcare, and energy [4]

Group 1 - The A-share market has been very active recently, particularly the computing power sector, which has shown strong performance since mid-August, with AI chips, AI servers, optical modules, and liquid cooling technology being standout hardware segments [1] - The strong performance of the computing power sector is driven by three main factors: first, the capital expenditure plans disclosed by major overseas cloud vendors exceeded expectations, maintaining high growth in global computing infrastructure investment; second, OpenAI's recent announcement of GPT-5 has attracted industry attention, with plans to double the scale of computing clusters in the next five months; third, feedback from the supply chain indicates robust demand for computing power, supported by new technological iterations that have spawned multiple high-growth sub-sectors [1] - Looking ahead, the demand for computing power is expected to grow exponentially due to breakthroughs in AI model performance, the evolution of multimodal technologies, and the rapid deployment of AI Agent applications, indicating a long-term high prosperity in the industry [1] Group 2 - Specific investment opportunities include GPU and ASIC chips, optical modules, optical chips, and fiber optic connectors, which are expected to continue benefiting from the growth in computing power [1] - Additionally, the rising demand for AI servers will lead to increased usage of PCBs, high-power power supply systems, and liquid cooling solutions, marking significant opportunities in these emerging fields [1]

Core Insights - The rapid development of China's computing power industry is highlighted, with significant growth in the computing power index and infrastructure [1][3] - The establishment of a comprehensive computing power platform across multiple provinces marks a new collaborative framework for computing power development in China [1][3] Summary by Sections Comprehensive Computing Power Development - As of June 2025, China's total number of operational computing center racks reached 10.85 million, with intelligent computing power scaling up to 788 EFLOPS [1] - The "2025 Comprehensive Computing Power Index" indicates a significant increase in storage capacity, projected to exceed 1680 EB by mid-2025, a 40% increase from 2023 [3] Computing Power Index - The index focuses on computing power scale and quality, with Hebei maintaining the top position for three consecutive years, while Zhejiang and Inner Mongolia showed significant improvements [4] - The eastern regions dominate in storage capacity, with Hebei and Guangdong exceeding 160 EB, while Jiangsu, Shanghai, and Beijing have established systems surpassing 100 EB [4] Network Capacity - Eastern provinces account for 70% of the top ten in network capacity, with the Yangtze River Delta region, particularly Zhejiang, Shanghai, and Jiangsu, leading the rankings [5] - The introduction of a new "model power" index assesses the comprehensive computing power development across provinces, highlighting the disparity between eastern and western regions [5] Ecological Collaboration - The establishment of the China Computing Power Platform aims to enhance compatibility and interoperability among various systems, facilitating a neutral and open service ecosystem [8] - As of July, over 1000 enterprises have registered on the platform, which has successfully managed the supply and demand of computing resources [8] Regional Development - The China Computing Power Platform in Shanxi is set to enhance resource management and monitoring capabilities, contributing to the province's computing power development [9] - The OISA ecological co-construction strategy aims to improve interconnectivity among GPU cards, promoting collaborative development in both intelligent and supercomputing sectors [10] Future Directions - The focus on enhancing data infrastructure is emphasized as a means to transition from a data-rich to a data-strong nation, seizing opportunities in the AI era [10]

Group 1 - The 2025 China Computing Power Conference opened with the theme "Building the Foundation of Computing Power to Lead the Future," highlighting the acceleration of the construction of China's computing power platform, with 10 provinces and cities already connected [1] - The China Computing Power Platform integrates computing resources from different regions and industries, creating a unified network for computing power scheduling, enabling comprehensive data collection and dynamic monitoring of computing power development [3] - The Ministry of Industry and Information Technology has issued the "Computing Power Interconnection Action Plan," aiming to establish a complete standard and rule system for computing power interconnection by 2026, and to achieve standardized interconnection of public computing power nationwide by 2028 [5] Group 2 - China's total computing power scale has been growing at an annual rate of around 30%, driven by the rapid development of artificial intelligence, leading to a surge in demand for intelligent computing power [7] - The China Telecom Research Institute predicts that by 2035, artificial intelligence will contribute over 11 trillion yuan to China's GDP, accounting for approximately 4% to 5% of GDP, potentially driving a tenfold to hundredfold increase in computing power demand [7] - The Ministry of Industry and Information Technology aims to guide the reasonable layout of intelligent computing facilities to improve the quality of intelligent computing supply, meeting the needs of emerging and future industries [9]

Core Viewpoint - The construction of China's computing power platform is accelerating, with multiple provinces and cities already connected, aiming to enhance resource utilization and promote intelligent transformation across industries [1][3][19]. Group 1: Development of Computing Power Platform - The computing power platform in China has integrated resources from various regions and industries, creating a unified scheduling network [6]. - Over 100 computing service providers have joined the platform, with more than 1,000 industry users registered, and nearly 100 mainstream foundational and vertical models connected [8]. - The Ministry of Industry and Information Technology has set goals for establishing a comprehensive interconnectivity standard and system by 2026, with a vision for a standardized computing power internet by 2028 [10]. Group 2: Growth and Demand for Intelligent Computing Power - The total scale of computing power in China has been growing at an annual rate of approximately 30%, driven by the rapid development of artificial intelligence [19]. - By 2035, artificial intelligence is projected to contribute over 11 trillion yuan to China's GDP, potentially increasing the demand for computing power tenfold or even a hundredfold [22]. - The intelligent computing power has been widely applied in various fields, including generative models, autonomous driving, smart cities, and industrial manufacturing, with over 23,000 innovative computing projects collected through application competitions [22]. Group 3: Future Projections - It is expected that by 2025, the scale of intelligent computing power in China will grow by more than 40% [22].

Group 1: China Computing Power Platform - The construction of the China Computing Power Platform is accelerating, with 10 provinces and cities officially connected to the computing power sub-platforms [1] - The platform aims to enhance cross-regional and cross-industry resource matching, improving the efficiency of computing resource utilization [1] - The platform has gathered over 100 computing service providers and registered more than 1,000 industry users, connecting nearly 100 mainstream foundational and vertical models [3] Group 2: Action Plans and Standards - The "Computing Power Interconnection Action Plan" aims to establish a complete standard and rule system for computing power interconnection by 2026, with a goal of achieving standardized interconnection by 2028 [5] - The plan envisions a computing internet capable of intelligent perception and real-time discovery [5] Group 3: AI and Computing Power Collaboration - The integration of computing power and artificial intelligence is accelerating the intelligent transformation of industries, with various applications showcased at the conference [6] - An AI drone countermeasure system demonstrated over 99% recognition accuracy and a response time of less than 3 seconds [8] - A computing service platform can dispatch one-sixth of the national computing power scale, supporting billions of computing calls daily [10] Group 4: Growth of Intelligent Computing Power - China's total computing power has been growing at an annual rate of around 30%, with intelligent computing demand rapidly increasing due to the development of AI [14] - By 2035, AI is expected to contribute over 11 trillion yuan to China's GDP, potentially driving a tenfold to hundredfold increase in computing power demand [14] - The intelligent computing power scale is projected to grow by over 40% in 2025, with applications in various sectors including generative models, autonomous driving, and smart cities [16]