华为智慧油气 解决方案 | • 勘探开发算力中心 02 | | --- | | • 油气勘探数据存储 05 | | • 智慧作业区 09 | | --- | | • 油气田一张网 21 | | • 华为星河 AI 高品质油气总部园区网络 25 | | • 数智化管网 30 | | --- | | • 管道一张网 36 | | • 智慧管网光通信 40 | 智慧勘探开发 / 智慧油气田 / 智慧管网 智慧勘探开发 01 华为智慧油气解决方案 勘探开发算力中心 背景与挑战 目前业界主流的勘探开发地学专业软件的计算硬件资源适配具有生态门槛，间接影响计算底座多元化发展与资 源共享。存算网资源未统一规划，资源云化进度慢、成本高。 无法跨集群完成大工区深度域地震数据处理、两宽一高海量数据和多工区大连片地震数据等作业任务，影响业 务推进效率。 多部门采购同类软件，单独部署，无法共享，软件利用率低，成本高。 解决方案简介 勘探开发算力中心解决方案基于华为云实现对算力资源的灵活调度，以及专业软件和数据的统一管理，支撑各 部门之间数据成果协同与共享，提升业务效率。 处理 解释 储层预测 地质建模 数值模拟 安全保障体系 运维保障 ...

Report Industry Investment Rating There is no information provided in the content about the report industry investment rating. Core Viewpoints of the Report - The development of Agentic AI is driving the transformation of enterprise ICT infrastructure and operations and maintenance, enabling enterprises to achieve digital and intelligent upgrades [10][11]. - Huawei is actively promoting the application of Agentic AI in enterprise operations and maintenance, providing a series of solutions and case studies to help enterprises improve operational efficiency and user experience [6]. - The future development of enterprise ICT infrastructure will focus on intelligent, autonomous, and collaborative operations and maintenance, with Agentic AI playing a key role [17][21]. Summary by Directory 01 Trend and AEI Overview 1.1 Agentic AI Accelerates Enterprise Digital and Intelligent Upgrade - The development of technology has gone through the information age, digital age, and is now moving towards the digital and intelligent age, with AI becoming the core driving force [9][10]. - Agentic AI is expected to reach the early majority adoption stage in 3 - 6 years, and by 2028, 60% of IT operations and maintenance tools will have AI agent functions [11]. 1.2 AI + New ICT Infrastructure is the Cornerstone of Enterprise Digital and Intelligent Transformation - Enterprises leading in digital and intelligent transformation focus on investing in digital infrastructure and using technology to create value [14]. - AI + ICT infrastructure innovation can accelerate the popularization and monetization of enterprise business, and Huawei predicts significant technological development by 2030 [15]. 1.3 Agentic AI Era of Enterprise ICT Infrastructure Operations and Maintenance - The development of the intelligent world has changed business applications, posing challenges to infrastructure and operations and maintenance systems [16]. - AEI, with features of Adaptive Multi - Agent, Autonomous O&M, and AI Native Infrastructure, aims to improve enterprise digital productivity [19]. - Agentic AI will reconstruct the enterprise operations and maintenance model, and Huawei is committed to promoting the transformation of operations and maintenance from automation to autonomy [21]. 02 Vision, Target Architecture, and Overall Plan 2.1 Huawei AEI Vision - Huawei aims to build a fusion - simplified Agentic AI operations and maintenance solution for enterprise data centers and smart campuses, enabling high availability, excellent experience, and simple operations [22]. 2.2 Huawei AEI Target Architecture - The operations and maintenance mode is transitioning from automation to autonomy, and Agentic AI provides technical support for autonomous operations and maintenance [24]. - The architecture includes intelligent infrastructure, intelligent data中台, intelligent business applications, and Agentic AI operations and maintenance systems, each with specific functions [26][27][30][31]. 2.3 Huawei AEI Overall Plan - Huawei offers three solution capabilities for data centers and smart campuses: computing - network - storage Agentic intelligent strong computing, enterprise - network - terminal Agentic intelligent auxiliary operations and maintenance, and ICT intelligent cloud management Agentic intelligent interconnection [35]. - The implementation of AEI is phased, with different stages achieving cross - domain self - intelligence and group intelligence [38][39]. 03 Value Scenarios and Solutions 3.1 Data Center - Data centers face challenges such as frequent failures, complex fault patterns, low resource utilization, and high energy consumption [42][43][44][45]. - The AEI@DC solution consists of three layers: intelligent infrastructure, Agentic AI operations and maintenance system, and business platform and application [52]. - The solution provides six operations and maintenance value scenarios, including fault handling, experience optimization, and efficient operations [54]. 3.2 Smart Campus - Different industries in smart campuses, such as education, healthcare, finance, and retail, face various operations and maintenance challenges [80][82][85][87]. - Huawei offers two solutions for smart campuses: AEI@Campus large - and medium - sized campus solution and AEI@Campus branch small - sized campus solution [89].

Group 1 - The rapid development of AI technology marks a new era in the technological revolution, indicating that the creation and application of knowledge are no longer solely human privileges [4] - Current AI applications are primarily focused on AI assistants with question-and-answer capabilities, which are often viewed as "black boxes" [4] - The potential for AI applications in industrial and service sectors remains largely untapped [4] Group 2 - The report "Intelligent World 2035" outlines a vision for AI development, exploring how technological integration will drive the transformation of industrial and service intelligent systems [5] - It highlights the potential applications of AI in various sectors, including healthcare, education, smart homes, smart cities, and business innovation [5] - The report discusses the synergistic effects of AI with other innovative technologies and the social and economic impacts of this transformation [5] Group 3 - Achieving the vision of AI development faces numerous technical challenges beyond general artificial intelligence [7] - The construction of intelligent systems is disrupting traditional systems engineering, requiring a combination of traditional ICT models and data-driven AI technologies [7] - A balance must be achieved between the correctness of system design and resilience during operation, with ongoing updates for continuous evolution [7] Group 4 - The vision presented in the report is broad and ambitious, contrasting with the approaches of tech giants that rely on machine learning and large-scale development [8] - Realizing this vision requires unprecedented technological breakthroughs and global collaboration [8] - The report emphasizes the importance of open ecosystems and international cooperation in advancing technology in the complex and transformative field of AI [9] Group 5 - The report identifies three key opportunities for the future of AI: more effective perception of the world, smarter model algorithms, and more efficient computing chips [11][12][13] - The integration of physical and digital worlds is essential for the evolution of general artificial intelligence (AGI) [11] - The development of new computing paradigms, such as quantum computing and neuromorphic chips, is crucial for achieving significant advancements in AI capabilities [13] Group 6 - The report outlines ten major technological trends that will shape the evolution of intelligent technology and the restructuring of social forms over the next decade [16] - These trends are interconnected and collectively form a complex ecosystem that supports the large-scale, real-time, and reliable interaction and decision-making of intelligent agents [17] - The report emphasizes the need for collaboration between technical experts and industry specialists to address the unique complexities of various sectors [18] Group 7 - The future intelligent world will require a sustainable energy foundation, with intelligent technology integrated into energy network management [18] - The report stresses the importance of ethical guidelines to ensure fairness, transparency, and accountability in AI algorithms [23] - The development of a dynamic regulatory framework for AI is essential to balance innovation and safety [23]

Investment Rating - The report does not explicitly state an investment rating for the intelligent campus industry Core Insights - The intelligent campus is defined as an all-intelligent, people-centric, green, and low-carbon self-evolving system that integrates physical, digital, and human spaces [113] - The report emphasizes the importance of digital transformation and the integration of advanced technologies such as AI, IoT, and big data in shaping the future of campuses [40][33] - The vision for intelligent campuses includes achieving zero-carbon status and promoting sustainable development through innovative energy management and green technologies [95][96] Summary by Sections Trends and Visions - The digital economy is driving the transformation of campuses, with projections indicating that by 2030, the digital economy will account for 30% of GDP [52] - Intelligent campuses will be characterized by being digital, converged, people-centric, resilient, and green [56] - The infrastructure of future campuses will feature 10 Gbps connections and digital platforms to support intelligent connectivity [57][58] Development Trends - Intelligent buildings will evolve to become core elements of campuses, enabling real-time connections and personalized services [61][62] - Data-based governance will become mainstream, enhancing campus management and operational efficiency [65][68] - Virtual-real fusion will integrate physical and digital experiences, creating new service scenarios [70][71] Scenarios - The report outlines ten typical scenarios for future intelligent campuses, including holographic AIOC, digital health services, and smart energy management [44] Key Technical Features - Six key technical features are proposed for future intelligent campuses: intelligent twins, spatial interaction, ubiquitous intelligent connectivity, intent-driven ultra-broadband, security and resilience, and all-domain zero carbon [33][24] Vision - The intelligent campus vision aims to create a sustainable and livable society by integrating connectivity and computing technologies [12][11] - The report highlights the need for ethical AI practices and cross-institutional data connections to foster collaborative innovation [13][14]

Investment Rating - The report does not explicitly provide an investment rating for the industry Core Insights - The report emphasizes the integration of new digital and intelligent technologies in the transportation sector, particularly in the development of smart highways, which are seen as a priority for many countries to enhance safety, efficiency, and user experience [16][17] - The report outlines a four-layer technical reference architecture for smart highways, which includes intelligent interaction, intelligent connection, intelligent central control, and smart applications, aimed at addressing challenges in safety, efficiency, and service [17][62] Summary by Sections Introduction - The introduction highlights the global trend towards digital and intelligent transformation in the transportation industry, driven by technologies such as cloud computing, big data, 5G, IoT, AI, and blockchain [16] - Smart highways are identified as a key area for development, aiming to improve the connection between vehicles, roads, and users to reduce accidents and enhance traffic efficiency [16][17] Current Status and Trends - Various countries are implementing long-term plans for smart highway development, with significant investments in technology research and application [19][20][21] - The report notes that while smart highway construction is progressing, standardization efforts are lagging behind, which poses challenges for cohesive implementation [24][25][26] - The integration of smart vehicles and intelligent roads is seen as essential for the future of transportation [29] Demand and Challenges - Smart highways are defined by their ability to integrate traditional road infrastructure with modern information technologies, focusing on spatial connectivity, information sharing, and resource integration [35] - The report identifies four common demands for smart highway construction: safety, efficiency, green development, and service quality [39] - Key challenges include low levels of information technology integration in road infrastructure, insufficient real-time data analysis capabilities, and the need for enhanced user experience [55][56][57] System Architecture - The report presents a comprehensive architecture for smart highways that emphasizes collaboration across cloud, network, edge, and terminal layers, aiming to solve business problems systematically [62][63] - The architecture is designed to facilitate real-time data interaction and decision-making, enhancing the overall efficiency and safety of transportation systems [63][64] Implementation Pathways - The implementation of smart highways is outlined in stages, starting from basic network perception to achieving a fully integrated smart network [74][75] - The report emphasizes the importance of building a digital twin platform and enhancing data sharing mechanisms to support smart transportation solutions [79] Key Technologies - The report discusses various technologies essential for smart highways, including software-defined cameras and millimeter-wave radar, which enhance data collection and analysis capabilities [81][82][86] - These technologies are crucial for improving traffic monitoring, safety, and operational efficiency in various weather conditions [86]

Investment Rating - The report does not explicitly state an investment rating for the industry Core Insights - The financial industry is undergoing a significant transformation with the migration of core systems to the cloud, driven by the need for digital transformation and the adoption of cloud-native technologies [7][8] - Major state-owned and joint-stock banks have completed pilot projects for migrating general business to core business on the cloud, entering the phase of large-scale migration [7] - The increasing reliance on cloud platforms for core business operations raises demands for high availability and reliability, as any service interruption can have severe social impacts [7][8] Summary by Sections 1. New Challenges Brought by Cloud Migration - Challenge 1: Designing high-availability cloud solutions and reliable operational support from an application perspective is critical as traditional methods are insufficient for core business needs [10][15] - Challenge 2: The rapid increase in the technology stack of cloud platforms complicates the implementation of end-to-end visibility and monitoring [10][17] - Challenge 3: The deep integration of cloud and network necessitates quick detection, localization, and recovery of issues [10][19] - Challenge 4: Addressing dual challenges of operational security and tenant security during cloud migration is essential [10][23] 2. Core Capabilities for Modernized Operations - The modernization of operations focuses on three main areas: platform operations, application operations, and security operations [31] - Platform operations must achieve full-link monitoring, deterministic fault recovery, and proactive risk governance [32][33] - Application operations should prioritize reliability from the design phase, utilizing high-availability designs to enhance application reliability [37] - Security operations require comprehensive control over operational processes to mitigate risks associated with operational errors and tenant security [38] 3. Detailed Operational Modernization Capabilities - Full-link monitoring involves constructing a comprehensive indicator system to quickly perceive faults across application and cloud platform layers [39] - Deterministic fault recovery is achieved through a combination of fault mode libraries and integrated cloud-network operations [64][65] - Proactive risk governance aims to anticipate and prevent risks through a systematic approach [36][68] 4. Application Observability and Monitoring - A layered design for observability is necessary, focusing on terminal, application, PaaS instance, and IaaS infrastructure layers [58] - The observability framework should integrate various metrics to provide a holistic view of application health and performance [60][61] - Simplified information aggregation is crucial for enhancing operational efficiency and fault handling [61][63]

Investment Rating - The report does not explicitly state an investment rating for the industry Core Insights - The financial industry is undergoing a significant transformation with the migration of core systems to the cloud, driven by the need for digital transformation and the adoption of cloud-native technologies [7][8] - Major state-owned and joint-stock banks have completed pilot projects for migrating general business to core business on the cloud, entering the phase of large-scale migration [7] - The increasing reliance on cloud platforms for core business operations raises demands for higher availability and reliability, as any service interruption can lead to severe social impacts and credit crises [7][8] Summary by Sections 1. New Challenges Brought by Cloud Migration - Challenge 1: Designing high-availability cloud solutions and reliable operational support from an application perspective is critical, as traditional methods cannot meet the new availability standards [15][16] - Challenge 2: The rapid increase in the technology stack of cloud platforms complicates the implementation of end-to-end monitoring [17][18] - Challenge 3: The deep integration of cloud and network requires quick detection, localization, and recovery of issues [19][20] - Challenge 4: Addressing dual challenges of operational security and tenant security during cloud migration is essential [23][24] 2. Core Capabilities for Modernized Operations - The modernization of operations focuses on three main areas: platform operations, application operations, and security operations [31] - Platform operations must achieve full-link monitoring and deterministic fault recovery to ensure the stability of core applications on the cloud [33][34] - Application operations should prioritize reliability from the design phase, utilizing high-availability designs to enhance application reliability [37] - Security operations need to implement comprehensive security controls throughout the operational process to mitigate risks associated with operational errors and tenant security [38] 3. Detailed Operational Modernization Strategies - Full-link monitoring should be established to cover all layers from applications to cloud platforms, ensuring quick fault detection and diagnosis [39][40] - Deterministic fault recovery mechanisms should be developed based on a fault mode library to facilitate precise fault diagnosis and rapid recovery [64][65] - A proactive risk governance approach is necessary to anticipate and prevent operational risks, leveraging digital and automated solutions [36][68]

Investment Rating - The report does not explicitly state an investment rating for the industry Core Insights - The report emphasizes the importance of innovative technologies in the AI and cloud computing sectors, particularly focusing on enhancing efficiency and performance in AI model training and deployment Summary by Sections Diversity Computing Scheduling - The report discusses the challenges of heterogeneous computing resource management, highlighting the need for unified scheduling of diverse computing resources such as CPU, GPU, and NPU to improve efficiency [17][22][23] - It introduces Huawei Cloud's diversity computing scheduling framework, which enhances distributed AI task scheduling and resource utilization through various innovative optimizations [22][23] Cloud-Edge Collaboration - The report outlines the increasing demand for real-time inference in industrial applications and the need for efficient deployment and operation of AI models [37][39] - Huawei's hybrid cloud solution supports centralized training and edge inference, enabling continuous model iteration and adaptation to changing environments [39][41] AI-Native Storage - The report identifies storage as a key bottleneck in AI model training efficiency, particularly with large-scale training clusters [51][55] - Huawei Cloud's AI-Native storage architecture addresses these challenges by providing high-performance data access and rapid checkpoint saving and recovery [52][55][63] Enhanced AI Network - The report highlights the significant communication overhead in AI model training, which can account for up to 40% of the training time [65][66] - It discusses the development of a lossless high-bandwidth network to optimize communication efficiency and reduce bottlenecks during training [69][78] Operator Acceleration - The report emphasizes the need for efficient tools and methodologies to enhance model performance and reduce the development threshold for operators [80][85] - Huawei's CANN heterogeneous computing framework aims to maximize hardware capabilities and streamline operator development processes [85][92] Full-Link Data Engineering - The report addresses the importance of data quality in AI model performance and the challenges in data acquisition and processing [97][100] - Huawei Cloud introduces a comprehensive data engineering framework with innovative tools to improve data quality and integration for AI model training [101][105]

Summary of Conference Call Records Industry Overview - The focus of the conference call is on the **autonomous driving** industry and its advancements, particularly influenced by **Deep-Seeking** technology and AI developments [1][2][3][4][5][6][7][10]. Key Points and Arguments 1. **Strategic Direction**: The company is committed to an "all-in" strategy for autonomous driving, emphasizing the significant role of AI and Deep-Seeking technology in this evolution [1][2]. 2. **Technological Advancements**: The progress in autonomous driving technology is closely linked to AI advancements, with capabilities improving at a rate of approximately **10 times annually** [2][3]. 3. **Cost Reduction**: The cost of autonomous driving systems has decreased significantly, dropping from over **50,000** to below **20,000**, representing a reduction of over **50%** [3][4]. 4. **Market Penetration**: The penetration rate of advanced autonomous driving features is expected to rise sharply, with projections indicating a jump from **10%** to **40-50%** by **2025** [4][5]. 5. **Comparative Analysis**: The domestic market is currently at a level just above VC1, with expectations to reach VC3, indicating substantial room for growth [3][4]. 6. **Industry Trends**: The overall industry is still in its early stages, with significant growth potential as more companies recognize the importance of advanced autonomous driving [5][6]. 7. **Performance Metrics**: The performance of leading players in the autonomous driving sector is expected to improve, with companies like BYD already implementing advanced driving features [5][6][12]. 8. **Investment Recommendations**: The company recommends investing in leading players in the smart vehicle sector, particularly those associated with Huawei and new energy vehicles [12][13][14][15]. Additional Important Insights 1. **Market Dynamics**: The call highlighted the increasing differentiation in the market, with some companies like BYD and Geely showing strong performance in the new energy vehicle segment [20][21]. 2. **Government Policies**: Local government policies are expected to stimulate demand further, particularly in the context of new energy vehicles [25][26]. 3. **Future Projections**: The overall outlook for the autonomous driving market is optimistic, with expectations for accelerated growth and increased market share for advanced features [21][22][27]. 4. **Technological Integration**: Companies are increasingly integrating advanced driving features into their models, which is expected to enhance competitiveness [48][49][50]. 5. **Financial Performance**: The financial performance of companies in the sector is projected to improve, with several companies expected to reach profitability by **2025** [46][49][55]. This summary encapsulates the key discussions and insights from the conference call, focusing on the autonomous driving industry's current state and future potential.

Investment Rating - The report does not explicitly state an investment rating for the industry or company. Core Insights - The AI large model is accelerating the intelligent transformation of various industries, with applications becoming increasingly widespread and enhancing operational efficiency and decision-making capabilities [19][21][25]. - The need for AI-Ready data infrastructure is emphasized, as it is crucial for supporting the training and deployment of AI models, ensuring high performance and data availability [20][22][34]. - The report highlights the importance of data quality and management, stating that 80% of the time in AI workflows is spent preparing high-quality data, which is essential for effective model training [36][40]. Summary by Sections AI Large Model Applications - AI large models are penetrating various sectors, achieving breakthroughs in coverage and precision, and enhancing generalization capabilities [21][25]. - The emergence of models like ChatGPT and advancements in video generation models signify a shift towards more complex AI applications [22][23]. Data Infrastructure Requirements - AI large models require robust data infrastructure characterized by high performance, strong consistency, and the ability to handle massive data volumes efficiently [13][17][34]. - The report defines "AI-Ready" data infrastructure as essential for the effective operation of AI models, focusing on aspects like scalability, flexibility, and real-time data access [11][12][34]. Challenges in Data Management - The report identifies significant challenges in data asset management, including data quality, standardization, and the prevalence of data silos, which hinder effective AI model training [36][40][41]. - Recommendations include establishing unified data management platforms and creating global AI data lakes to enhance data accessibility and quality [39][41]. Industry-Specific Applications - Various industries, including finance, healthcare, and manufacturing, are leveraging AI large models for applications such as risk assessment, personalized services, and operational efficiency [25][26][32]. - The report outlines specific use cases in banking, healthcare, and public services, demonstrating the transformative potential of AI technologies across sectors [25][26][32].