Core Viewpoint - Huawei introduces its latest SuperPoD products at MWC Barcelona 2026, aiming to establish a resilient computing foundation and create new opportunities globally through open-source collaboration and technological innovation [1]. Group 1: Product Launch and Features - Huawei showcases the Atlas 950 SuperPoD, TaiShan 950 SuperPoD, and a series of computing solutions designed to meet the increasing demand for high-performance computing in AI applications [1]. - The Atlas 950 SuperPoD connects up to 8192 NPUs via UnifiedBus, providing ultra-high bandwidth, ultra-low latency, and unified memory addressing, functioning as a single logical unit for learning, reasoning, and information processing [1]. - The TaiShan 950 SuperPoD is highlighted as the first universal computing SuperPoD in the industry, alongside next-generation servers like TaiShan 500 and TaiShan 200, offering flexible computing options for workloads ranging from high to low intensity [1]. Group 2: Open Source and Ecosystem Development - Huawei emphasizes its commitment to open-source software and systems, aiming to accelerate developer innovation and ecosystem prosperity [1]. - The company plays a key role in the development of openEuler, which has rapidly become one of the leading global open-source operating system communities [1]. - Huawei has made its heterogeneous computing architecture CANN fully accessible, allowing all software components to be openly available for developers, thus supporting various open-source communities and projects [1].

Core Insights - Huawei has rapidly evolved from using open-source software to becoming a major contributor to various large open-source projects since 2010, with over 6,000 employees involved in development [1][3]. Group 1: Open Source Contributions - Huawei is a top player in the global open-source field, being a founding member of several prominent international open-source foundations and contributing core code to many communities [3]. - The company has initiated over ten significant open-source projects, particularly in foundational software, which has garnered widespread support from global developers [3]. Group 2: AI and Computing Frameworks - Huawei's CANN architecture, launched in 2019, facilitates AI developers in utilizing underlying computing power and is set to be fully open-sourced by 2025, allowing clients to optimize their usage [4]. - The CANN community is actively collaborating with universities to cultivate professional talent, enhancing the AI ecosystem [4]. Group 3: Hardware and Software Ecosystem - Huawei's Kunpeng processor, launched in 2019, has made significant strides in supporting major open-source software, addressing the challenges posed by the dominance of X86 architecture [5]. - The company has developed the Kunpeng DevKit and BoostKit to improve computing performance through software-hardware synergy, boosting the Kunpeng ecosystem's attractiveness [5]. Group 4: Operating Systems and Databases - The openEuler operating system, based on Linux, has attracted over 2,100 enterprises and institutions, with more than 26,000 contributors, and is projected to exceed 16 million installations by the end of 2025 [6]. - Huawei's openGauss project, a relational database, is gaining traction in critical industries and will continue to enhance its capabilities to support distributed architectures and multi-modal data [7]. Group 5: Strategic Vision - Huawei's strategy in the computing sector focuses on hardware openness, software open-sourcing, enabling partners, and talent development to drive innovation in China's computing industry [7].

Core Insights - The emergence of SuperPods as a new AI computing infrastructure has become a focal point in the industry since 2025, with Huawei's Ascend 384 SuperPod leading the way in performance metrics compared to foreign competitors [1][3] - The demand for computing power is far from being met, with token consumption expected to exceed trillions daily in China, highlighting the inadequacy of simply stacking servers to address the computing gap [3][4] Group 1: SuperPod Characteristics - SuperPods are not merely about stacking chips; they represent a fundamental restructuring of traditional computing architectures, enabling equal interconnectivity among CPUs, NPUs, and memory units [4][6] - Key features of a true SuperPod include high bandwidth to eliminate communication delays, low latency, and the ability to form a logically unified system through unified memory addressing [6][7] Group 2: Efficiency and Performance - SuperPods can significantly enhance computing efficiency, with potential model utilization rates increasing from 30% to 45%, effectively a 50% improvement, which can help mitigate the limitations of chip manufacturing processes [7][8] - The architecture of SuperPods differs from traditional systems, as Huawei employs optical communication technology, allowing for greater scalability and interconnectivity compared to NVIDIA's copper-based systems [8][9] Group 3: Innovation and Ecosystem - Huawei's systematic innovation in chip design, optical components, and foundational protocols has positioned it uniquely in the market, leveraging over 20 years of experience in optical technology [9][12] - The company is also developing general computing SuperPods, with the TaiShan 950 SuperPod set to launch in Q1 2026, aimed at replacing various server applications [11][12] Group 4: Software and Community Engagement - The success of SuperPods relies not only on hardware but also on a robust software ecosystem, including open-source initiatives like CANN and openEuler, which are crucial for fostering industry collaboration [14] - Huawei has engaged a large developer community, with 3.8 million registered developers for Kunpeng and nearly 4 million for Ascend, emphasizing the importance of open-source collaboration in the AI era [14]

Core Viewpoint - The emergence of SuperPod as a new AI computing infrastructure has become a focal point in the industry since 2025, with Huawei's Ascend 384 SuperPod leading the way in performance metrics compared to foreign competitors [1][3]. Group 1: SuperPod Concept and Advantages - SuperPod is not merely about stacking chips; it represents a fundamental restructuring of traditional computing architecture, enhancing communication efficiency among CPU, NPU, and memory units [4][6]. - The key advantages of SuperPod over traditional clusters include significantly improved computational efficiency, with potential model computing utilization rates increasing from 30% to 45%, equating to a 50% performance boost [7][8]. Group 2: Technical Challenges and Innovations - Building a true SuperPod is complex; Huawei's Ascend 384 SuperPod consists of 12 computing cabinets and 4 bus cabinets, while NVIDIA's NVL72 system is confined to a single cabinet due to architectural differences [8]. - Huawei employs optical communication technology for interconnection, allowing for greater scalability beyond single cabinet limitations, while traditional systems face constraints with electrical signal transmission [8][9]. Group 3: Systematic Innovation and Ecosystem Development - Huawei's systematic innovation includes proprietary chip development, optical device capabilities, and foundational protocols, enabling the creation of SuperPods that leverage full optical interconnectivity [9][12]. - The company is also developing general computing SuperPods, such as the TaiShan 950, which aims to replace various server applications by 2026 [9][11]. - A robust software ecosystem, including open-source initiatives like CANN and openEuler, is essential for the operation of SuperPods, with a focus on collaborative development within the industry [14].

Core Insights - The article emphasizes the importance of open-source initiatives in driving innovation and collaboration within the industry, highlighting Huawei's evolution from using and participating in open-source to leading it since 2010 [1][2] Group 1: Open Source Strategy - Huawei's computing industry has adhered to the principle of "hardware openness, software open-source, enabling partners, and developing talent" since its establishment in 2019, which is seen as a core strategy rather than just a slogan [2] - The company aims to provide not only usable but also high-performance products, leveraging open-source to foster industry collaboration and innovation [2] - Huawei has registered 3.8 million developers in the Kunpeng ecosystem and nearly 4 million in the Ascend ecosystem, collaborating with 9,800 ISV partners to create over 20,000 industry solutions across various sectors [2][3] Group 2: Open Source in General and Intelligent Computing - In the general computing domain, Huawei fills gaps in domestic software through open-source projects like openEuler and openGauss, enhancing collaboration among developers to optimize foundational software [3] - Huawei's open-source journey has evolved through three stages: starting with using open-source, then participating in it, and finally leading it by launching significant projects [3] - The open-source ecosystem significantly reduces the cost of developing foundational software for small and medium enterprises by over 80%, enhancing overall industry efficiency [3] Group 3: Empowering Industry Collaboration - In the intelligent computing sector, the demand for computing power is surging due to the increasing scale of large models and data requirements [4] - Huawei's open-source strategy for CANN aims to address industry pain points by allowing developers to optimize operators based on their specific scenarios, thereby improving computing efficiency [5] - The CANN framework serves as a bridge between hardware and upper-level development tools, facilitating collaboration and innovation in the AI era [5][6] Group 4: Supernode Technology and Open Source - Huawei has fully open-sourced its Lingqu protocol, which is essential for high-bandwidth, low-latency interconnections in supernode operations, aiming to build a complete industry chain in China [6] - The company emphasizes that supernode technology, combined with open-source initiatives, acts as a dual engine for advancing the computing industry [6] - Key features of supernode technology include high bandwidth, low latency, and unified memory addressing, which are crucial for efficient data exchange and overall performance [6]

Core Viewpoint - NVIDIA currently holds a near-monopoly in the AI training and inference chip market, driven by advanced technology and an unmatched ecosystem, making it the highest-valued public company globally with a market capitalization of approximately $4.5 trillion as of November 2025, and a year-over-year revenue growth of about 62% in Q3 2025 [2]. Competitive Landscape - NVIDIA faces competition from traditional chip giants like AMD and Intel, as well as tech companies like Google and Amazon with their custom chips, and emerging players like Cerebras and Groq. However, none have significantly challenged NVIDIA's leadership position so far [2]. - The AI compute chip market has two main applications: training and inference, with training being the core bottleneck in the early and mid-stages of large model development [4][5]. Training Dominance - NVIDIA's dominance in training compute stems from advanced technology and a monopolistic ecosystem. The training of large models requires massive computational power, necessitating large-scale chip clusters and a comprehensive software system to connect engineers, chips, and models [6]. - Key requirements for training chips include single-chip performance, interconnect capabilities, and software ecosystem [6]. - NVIDIA excels in single-chip performance, but competitors like AMD are closing the gap. However, this alone does not threaten NVIDIA's lead in AI training [7]. - Interconnect capabilities are crucial for large model training, with NVIDIA's proprietary NVLink and NVSwitch enabling efficient interconnectivity at a scale of tens of thousands of chips, while competitors struggle to achieve similar scales [7]. Ecosystem Advantage - NVIDIA's ecosystem advantage is primarily software-based, with CUDA being a well-established programming platform that fosters a strong developer community and extensive resources, enhancing user stickiness [8][9]. - The ecosystem's network effects mean that as more developers engage with CUDA, its value increases, creating a significant barrier to entry for competitors [10]. Inference Market Dynamics - Inference requires significantly fewer chips than training, leading to reduced interconnect demands. Consequently, NVIDIA's ecosystem advantage is less pronounced in inference compared to training [12]. - Despite this, NVIDIA still holds over 70% of the inference market share due to its competitive performance, price, and development costs [13]. Challenges to NVIDIA - Competitors must overcome both technical and ecosystem challenges to compete with NVIDIA. If they cannot avoid ecosystem disadvantages, they must achieve significant technological advancements [15]. - In the U.S., challengers are focusing on custom AI chips (ASICs), with Google's TPU showing promising results. However, the ecological disadvantage remains a significant hurdle [16]. - In China, U.S. export restrictions on advanced chips have created a protected market, limiting NVIDIA's ecosystem influence and presenting opportunities for local chip manufacturers [17][18]. Strategic Considerations - The geopolitical landscape has led to a potential rise of strong domestic competitors in China, as developers begin to adapt to local ecosystems like CANN, despite initial challenges [19]. - The U.S. government's recent policy shift allowing NVIDIA to sell advanced chips to China under specific conditions reflects a recognition of the need to maintain NVIDIA's competitive edge while managing technological disparities [19]. - A balanced approach is necessary for China to foster its AI chip industry while allowing for essential imports to support core AI projects [19].

Core Viewpoint - The article discusses Huawei's recent announcement regarding the open-source of its Ascend CANN software, which aims to lower the barriers for AI tool development and foster a new AI computing ecosystem [2][30]. Group 1: CANN Open Source and Developer Empowerment - CANN, which stands for Compute Architecture for Neural Networks, serves as a bridge between AI training frameworks and underlying AI chips, allowing developers to utilize computing power without needing to understand chip details [2][5]. - The open-source nature of CANN has garnered significant attention in the industry, as it empowers developers to define computing capabilities and customize their AI models [2][6]. - CANN supports seamless integration with major AI frameworks such as PyTorch, TensorFlow, MindSpore, and PaddlePaddle, enhancing developer flexibility [5][6]. Group 2: Development Paths Offered by CANN - CANN provides three development paths for different types of developers: 1. For those familiar with Python, CANN integrates with the Triton ecosystem, allowing easy migration of existing code [9]. 2. For system-level programmers seeking high performance, Ascend C offers low-level resource management capabilities [10]. 3. For developers looking for ease of use, the CATLASS operator template library simplifies the creation of matrix multiplication operators [11][13]. - The MLAPO fusion operator, part of the CATLASS library, significantly reduces computation time and enhances performance in large models [15]. Group 3: Architectural Innovations - CANN's architecture features a layered decoupling approach, allowing independent evolution of components, which reduces integration complexity for developers [21][22]. - The decoupling enables developers to selectively upgrade specific components based on their needs, facilitating easier customization and integration [23][29]. - CANN has transitioned from a monolithic software structure to a modular one, with independent components for various functionalities, enhancing flexibility and performance [24][26]. Group 4: Open Source Community and Growth - The open-source initiative of CANN is actively progressing, with over 27 sub-projects and a total of more than 3,700 stars on its repositories [35]. - The community-driven approach invites developers to contribute, thereby expanding the ecosystem and enhancing the technology's value through collaborative efforts [31][32]. - CANN's repositories include a variety of core libraries and tools, providing developers with ready-to-use resources for AI application development [16][36].

Core Insights - The launch of the Ascend 384 super node computing cluster in Shenzhen marks a significant advancement in AI computing capabilities, transitioning from traditional CPU-based systems to a more integrated and efficient architecture [1][2]. Group 1: Project Overview - The Ascend 384 super node is a collaborative project between Huawei and the Shenzhen Longgang District Urban Investment Group, designed to enhance AI computing power [1]. - This project represents a shift in AI computing clusters from "patchwork integration" to "system fusion," addressing issues like high communication latency and low collaboration efficiency [1]. Group 2: Technical Advancements - The new cluster features 384 high-performance NPUs and integrates 192 of the latest high-performance Kunpeng CPUs, achieving a threefold increase in training performance and a fourfold increase in inference performance [1][2]. - The architecture allows any processor within the cluster to access data center resources as if they were local memory, significantly improving overall system performance [1]. Group 3: Strategic Implications - The deployment of the Ascend 384 super node is a key initiative in the Longgang District's strategy to develop a comprehensive AI resource pool, aiming to provide robust support across various sectors such as government, finance, energy, and manufacturing [2]. - The project aligns with Longgang's "All in AI" strategy, which includes initiatives like the CANN open-source strategy and the establishment of a data company to leverage extensive data resources [2]. Group 4: Ecosystem Development - The collaboration between CANN as the AI computing engine and the open-source HarmonyOS as the operating system aims to create a seamless flow of computing resources across multiple devices [3]. - This integrated approach is expected to attract global AI algorithm developers and HarmonyOS application developers to Shenzhen, fostering a collaborative technology ecosystem and accelerating the development of the AI industry in the Greater Bay Area [3].

Core Insights - The core theme of the forum is "Open Drives Innovation," emphasizing the shift from competition to collaboration in the realm of intellectual property and innovation [1][10][12] Group 1: Forum Overview - The sixth Huawei Innovation and Intellectual Property Forum gathered representatives from various international organizations and companies to discuss the role of intellectual property in fostering collaborative innovation [1][2] - Huawei's Chief Legal Officer, Song Liuping, highlighted that the essence of intellectual property is not exclusivity but rather the orderly and efficient dissemination of innovative results [2][4] Group 2: Huawei's Contributions - As of the end of 2024, Huawei has obtained over 150,000 valid patents globally, with R&D investment exceeding 20% of annual revenue, totaling over 1.2 trillion yuan in the past decade [4][7] - The forum showcased Huawei's "Top Ten Inventions" for 2024, which include significant technological advancements such as the Scale Up ultra-large-scale computing platform and the HarmonyOS full-stack architecture [4][6] Group 3: Knowledge Sharing and Collaboration - The upgraded "Chasi Patents" platform was introduced, enhancing patent search and analysis capabilities, thereby accelerating knowledge flow and innovation [6][9] - Huawei's commitment to open innovation is reflected in its extensive participation in global ICT standards and its collaboration with international licensing platforms, generating over $630 million in patent licensing revenue in 2024 [8][9] Group 4: Open Innovation Strategy - Huawei's strategy emphasizes that knowledge sharing enhances social value rather than diminishing rights, promoting a cycle where patent protection leads to commercial returns that fund further R&D [9][10] - The forum underscored the importance of establishing standardized interfaces and shared platforms to make innovation more efficient and inclusive [10][12] Group 5: Future Implications - The discussions at the forum suggest that open innovation is becoming a dominant theme in global technological collaboration, with the potential to significantly enhance cross-industry integration [10][12] - The evolving landscape of technology, including AI and quantum computing, presents challenges in establishing sustainable cooperation mechanisms, highlighting the need for intellectual property to serve as a bridge rather than a barrier [10][12]

Core Viewpoint - The ultimate competition in AI between China and the US may not be about computing power but rather about electricity supply, as AI's increasing demands for power will determine who can sustain their operations longer [5][10]. Group 1: US AI Industry Challenges - The bottleneck for the US AI industry has shifted from chip availability to electricity supply, with major tech companies struggling to secure sufficient power for their operations [9][10]. - AI applications, such as ChatGPT, consume significant amounts of electricity, with estimates indicating that daily operations could power 17,000 American households [11]. - The aging US power grid, built decades ago, is unable to meet the surging electricity demands, leading to potential power shortages in certain regions by 2030 [11][12]. Group 2: China's Energy Advantage - China is leveraging its energy advantages to compensate for product performance disadvantages in the AI sector, particularly through initiatives like "East Data, West Computing" [15][18]. - This initiative aims to transfer data processing from energy-rich western regions to eastern areas where demand is high, optimizing the use of clean and inexpensive electricity [16][18]. - The recent electricity subsidy for AI data centers in China is seen as a strategic move to enhance competitiveness against US firms by reducing operational costs [5][7][25]. Group 3: Cost Structure and Subsidies - The electricity subsidy fundamentally alters the cost structure for companies considering domestic chips, making it more attractive to switch from foreign to local alternatives [25][32]. - The new cost formula for AI operations in China now includes reductions in energy costs due to the "energy scheduling dividend" and fiscal subsidies, making domestic chips more competitive [31][32]. - The subsidy is designed to cover the ecological migration costs associated with transitioning to domestic chips, providing critical time for the development of local ecosystems [32]. Group 4: Strategic Implications - The combination of energy advantages, subsidies, and domestic chip development forms a closed-loop system aimed at challenging Nvidia's dominance in the AI ecosystem [33][40]. - As AI competition evolves from a focus on chip performance to resource management, China's ability to maintain lower operational costs could provide a significant competitive edge [41][45]. - The strategic focus is not only on defense but also on offensive maneuvers against established players like Nvidia, with the goal of redefining the competitive landscape [42][46].