格隆汇10月28日｜思泉新材(301489.SZ)今日股价收跌16.98%报203.3元，全天成交额放大至近30亿元。 同行业的锦富技术官微消息显示，锦富技术定制开发的0.08毫米铲齿散热架构已获得某台湾客户的订 单，已用于B200芯片的液冷散热系统。此外，针对下一代B300芯片的适配方案也已完成多轮送样测 试，反馈良好，进入生产准备阶段。有业内人士称，市场的表现一定程度上是在反应思泉新材"业务被 抢"。 ...

Group 1 - The development and application demand for AI technology is driving the continuous increase in market requirements for GPU performance, leading to accelerated iterations and upgrades of GPU chips [1] - Current GPU products are evolving from the B200 to the new generation B300, both based on the Blackwell architecture; GB200 and GB300 represent the core development direction of data center computing power [1] Group 2 - The significant increase in chip power and computing performance has made heat dissipation a key bottleneck in performance release [2] - Jinfu Technology has developed a 0.08mm serrated heat dissipation architecture, which has received orders from a Taiwanese customer for use in the liquid cooling system of the B200 chip, utilizing the latest MLCP technology to effectively address TDP thermal effects for processors with power consumption of 1800W-2000W and above [2] - Jinfu Technology is deepening technical cooperation with leading global GPU companies and their ODM partners, aiming to complete reliability verification before large-scale shipments of GB300 and enhance R&D investment in microchannel cooling plate architecture [2]

Core Insights - Shenzhen Cheng Tian Wei Ye Technology Co., Ltd. has established a joint venture, SuperX Cooltech, with SuperX AI Technology Limited to focus on AI server and high-performance computing liquid cooling systems [1][2] - The partnership signifies a strategic move in Cheng Tian Wei Ye's global strategy for digital and energy thermal management [1][2] Group 1: Liquid Cooling Technology and Market Position - Liquid cooling technology is becoming essential for AI data centers due to the limitations of traditional air cooling methods, especially with the rise of high-power GPUs [2] - The joint venture aims to provide efficient and reliable liquid cooling solutions for the global market, excluding mainland China and Hong Kong [1][2] - Cheng Tian Wei Ye is leveraging its expertise in semiconductor materials and thermal management to create a comprehensive liquid cooling product line [2][5] Group 2: Core Product System of the Joint Venture - The joint venture will develop a core product system that includes high-strength nano-injection liquid cooling plates, microchannel liquid cooling plates (MLCP), liquid cooling modules, and CDU systems [3][4] - The MLCP is designed for high-density cooling solutions, supporting precise temperature control and thermal load balancing [3][4] - The integrated solutions aim to achieve system-level energy efficiency with a PUE of less than 1.1 [3] Group 3: Key Technological Innovations - The joint venture will utilize advanced technologies such as nano-injection molding, material reduction techniques, fluid optimization designs, and intelligent control systems [4] - These innovations are expected to provide a competitive edge in the liquid cooling market [4] Group 4: Global Market Outlook - The global liquid cooling systems market is projected to grow significantly, driven by increasing power density and efficiency demands in data centers [5] - The MLCP technology is anticipated to enter mass production by mid-2026, with strong interest from major AI chip customers [5] - The joint venture's product offerings are positioned to cover the entire thermal management chain from chip to cabinet, enhancing its market potential [5]

Core Insights - The article discusses the increasing demand for AI computing power and the challenges in thermal management, particularly the introduction of microchannel liquid cooling plate (MLCP) technology as a potential solution [2][57]. Group 1: MLCP Technology Overview - MLCP technology integrates a metal cover and liquid cooling plate with a microchannel design, enhancing cooling efficiency and reducing system size [4][6]. - The cost of MLCP is projected to be 3 to 5 times higher than existing cooling solutions, with manufacturing costs for GPUs potentially increasing by 5 to 7 times if MLCP is widely adopted [2][4]. Group 2: Thermal Management Challenges - The thermal power of the Rubin GPU is expected to rise from 1.8 kW to 2.3 kW, necessitating the introduction of MLCP technology by late 2026 [7][9]. - Current cooling solutions may not suffice for the increasing power demands of AI chips, indicating a critical need for advanced thermal management technologies [2][9]. Group 3: Industry Developments - Boyd, a leading supplier, has delivered 5 million liquid cooling plates to large-scale data centers, highlighting the growing importance of liquid cooling technology in AI computing [10]. - The article emphasizes that MLCP technology is gaining traction in the industry, with ongoing research and development efforts to address its challenges [6][57]. Group 4: Future Outlook - The article concludes that MLCP technology represents a significant advancement in thermal management for high-power AI platforms, despite facing production and technical validation challenges [57][59]. - The upcoming Thermal Management Industry Conference will focus on the latest developments in MLCP and liquid cooling technologies, providing a platform for industry experts to discuss solutions to the growing thermal challenges in AI computing [59].