Core Viewpoint - The Ministry of Industry and Information Technology (MIIT) has announced the construction of a national computing power interconnection node system, which is expected to enhance the efficiency and service level of public computing resources and promote high-quality development in the computing power industry [2][3]. Group 1: National Computing Power Interconnection Node System - The MIIT's notification outlines a "1+M+N" system for national computing power interconnection nodes, aiming to standardize and facilitate the efficient flow of computing resources across different regions and sectors [2][3]. - The construction goals include establishing a supply-demand matching system for computing power in regions and key industries, enhancing overall computing power levels [2][3]. - Regional nodes will provide a unified service platform for interconnection, including services such as resource aggregation, selection, and monitoring [2][3]. Group 2: Industry Trends and Demand - The domestic computing power demand is expected to accelerate, driven by rapid iterations of domestic AI models and the evolving interaction methods that are shaping the demand for inference-side computing power [4][5]. - Major AI models from companies like Alibaba and Baidu are being released, which will further drive the commercialization of these models and increase the demand for computing power [5][6]. - The infrastructure for AI computing power is undergoing systematic upgrades to support the explosive demand for training and inference, transitioning from vertical scaling to distributed cluster expansion and cross-domain collaboration [6][7]. Group 3: Hardware Innovations - Key hardware components such as optical modules and switches are innovating around high speed, low power consumption, and low cost, with a focus on co-packaged optics (CPO) and linear drive pluggable optics (LPO) [7]. - Domestic optical module manufacturers are gaining a leading position in the global market, which positions them to benefit significantly from the expansion of AI computing infrastructure [7].

Investment Rating - The industry investment rating is "Positive" (maintained) [1] Core Insights - The domestic model release period is intensifying, with significant models such as DeepSeek's open-source OCR2, Kimi's K2.5, Alibaba's Qwen3-Max-Thinking, and Baidu's Wenxin 5.0 being launched recently. ByteDance plans to release three new AI models in February, indicating a rapid acceleration in model commercialization [5] - Domestic cloud vendors are maintaining high capital expenditures, with ByteDance planning 160 billion yuan for 2026, up from approximately 150 billion yuan in 2025. Alibaba is also advancing a 3-year plan for 380 billion yuan in AI infrastructure [5] - The rapid iteration of domestic models is expected to significantly increase demand for inference-side computing power, with 2026 being a pivotal year for the deployment of domestic supernodes. Major companies like Huawei and Alibaba are launching new supernode solutions [5] - Investment suggestions highlight that the acceleration of domestic model iterations and improvements in inference-side performance will benefit the domestic computing power industry chain, with a focus on companies like Chipone Technology, Huafeng Technology, and Weicai Technology [5] Summary by Sections Recent Market Performance - The report notes a recent market performance with fluctuations, including a -13% change in one segment and a 67% increase in another [2] Key Company Ratings - Chipone Technology: Market cap of 109.3 billion yuan, with a rating of "Increase" [4] - Huafeng Technology: Market cap of 42.1 billion yuan, with a rating of "Increase" [4] - Weicai Technology: Market cap of 17.9 billion yuan, with no specific rating provided [4]

Core Insights - The report from CITIC Securities indicates that the supernode solution is expected to scale rapidly, serving as a fundamental computing unit for future AI infrastructure, with advantages such as efficient communication bandwidth and native memory semantics [1][2] Group 1: Supernode Development - The MoE (Mixture of Experts) architecture imposes new hardware requirements, leading to the emergence of scale-up supernodes [2] - Supernodes face complex systemic challenges compared to traditional eight-card servers, including heat dissipation, stability issues from mixed optical and copper interconnects, and long-term reliability concerns [2][3] - The current phase of supernode solutions is characterized by a variety of competing technologies, with domestic solutions like Huawei's CloudMatrix384 and Alibaba's Panjiu emerging [3] Group 2: Technical Challenges and Solutions - As computing density increases, liquid cooling solutions with a PUE closer to 1, such as phase change immersion cooling, may see greater development opportunities if stability issues can be resolved [4] - The complexity of supernode servers has significantly increased, requiring deep consideration of chip integration, heat dissipation, and interconnects, transforming server manufacturers into core system integrators [5] Group 3: Investment Strategy - The supernode technology is in its early stages, with the MoE architecture likely to become mainstream, presenting new adaptive requirements for hardware development [7] - The report suggests that companies with customization capabilities and supply chain management skills in the server manufacturing sector are likely to see significant growth opportunities [7]