Investment Rating - The report maintains a "Positive" investment rating for the industry [5] Core Viewpoints - The CXL solution optimizes AI storage architecture, and leading manufacturers are expected to accelerate its application [3][10] - The CXL memory pooling solution can significantly enhance storage system efficiency and reshape the memory hardware composition in AI computing facilities [8][10] - The demand for memory capacity is increasing due to AI inference processes, and current server memory upgrades are constrained by slot numbers and single memory stick capacities [10][19] Summary by Sections 1. CXL Solution Optimizes Storage Efficiency and Adapts to AI Inference Needs - The CXL solution helps expand memory capacity and optimize storage architecture, addressing limitations in existing AI computing facilities [19] - CXL memory pooling can achieve resource integration and unified scheduling, supporting larger-scale and higher-concurrency model training and inference tasks [21][24] - CXL technology is expected to significantly reduce the total cost of ownership (TCO) for data center systems by optimizing memory configurations [45][46] 2. CXL-Related Hardware and Software Are Gradually Improving, with Leading Manufacturers Accelerating Application - CXL specifications are continuously upgraded, with transmission performance increasing from 32 GT/s to 128 GT/s by 2025 [49][50] - Major manufacturers, including NVIDIA and Alibaba Cloud, are accelerating their CXL solution deployments [58][65] 3. CXL Solution Penetration Rate Is Expected to Continue Rising, Opening Growth Space for the Industry - The total share of CXL in server DRAM is projected to grow from nearly zero in 2024 to about 15% by 2030 [70] - The proportion of servers capable of supporting CXL functionality is expected to reach 68% by 2026 and 99% by 2030 [72] 4. CXL Applications Are Expected to Accelerate, with Related Companies Deeply Benefiting - Key investment targets include: - **Lanke Technology**: Rapid revenue growth, with 2025 revenue reaching 5.46 billion yuan, a 50% year-on-year increase [76][79] - **Jucheng Co., Ltd.**: Revenue of 1.22 billion yuan in 2025, a historical high, with a 25% year-on-year increase [86] - **Jiangbolong**: Released CXL 2.0 memory expansion module, with 2024 revenue of 17.46 billion yuan, a 72% year-on-year increase [91][93]

Summary of CXL Industry and Company Insights Industry Overview - **CXL Protocol**: Aims to address the mismatch between CPU/GPU processing speeds and memory bandwidth, facilitating memory pooling and sharing to support AI model training and inference [1][2] - **Market Size**: The CXL market is projected to reach $60 billion by 2030, with the CXL Switch chip market expected to grow to $730 million [1][10] - **Growth Drivers**: The demand for AI servers is significantly higher than for general servers, with storage needs increasing several times, necessitating advanced memory solutions [1][6] Core Insights - **CXL Technology**: Utilizes PCIe as a physical layer to enable memory pooling, with three main sub-protocols: CXL.io, CXL.cache, and CXL.mem, which enhance device interconnectivity and memory resource sharing [2] - **Application Scenarios**: CXL is applicable in three main areas: dedicated accelerator cards without memory, accelerator cards with local memory, and memory expansion for servers [3] - **Ecosystem Participants**: The CXL Alliance includes major players like Intel, AMD, NVIDIA, Google, and Dell, driving the protocol's evolution since its introduction in 2019 [5] Technical Developments - **Version Evolution**: CXL has rapidly evolved from version 1.0 to 2.1, with version 4.0 planned for 2025, reflecting the urgent market demand for high bandwidth and storage density solutions [5][4] - **Memory Pooling**: CXL's memory pooling technology is crucial for expanding server memory capacity, with ongoing exploration of hybrid memory solutions combining DRAM and NAND Flash [7] Market Dynamics - **CXL Switch Chip**: Essential for transitioning from memory expansion to pooling, with a projected market size of $730 million by 2030 [10] - **Key Components**: The CXL memory expansion system's core components include CXL Controllers and CXL Switch chips, with significant contributions from companies like 澜起科技 (Lanqi Technology) and Astera Labs [9][11] Company Insights - **澜起科技 (Lanqi Technology)**: Holds approximately 40% market share in the memory interface sector, with a strategic focus on CXL expansion chips and other related technologies. Expected to achieve a profit of 4 billion yuan by 2026 due to the growing server market [11] - **Investment Opportunities**: In addition to 澜起科技, investors are advised to consider companies like Astera Labs and Rambus in the US market, which are positioned to benefit from the CXL-driven industry transformation [12] Additional Considerations - **Light Communication Integration**: The combination of CXL technology with optical communication is anticipated to enhance bandwidth and reduce latency, further driving market growth [8] - **Future Trends**: The ongoing evolution of CXL technology is expected to meet the increasing storage demands of AI applications, making it a critical area for investment and development [6][8]

Core Viewpoint - The stock of 澜起科技 (Lianqi Technology) surged over 8% to a new high of 188 HKD, driven by positive market sentiment regarding the growth in CPU server demand and advancements in AI applications, which are expected to boost the company's memory interface business and profitability in the coming years [1] Group 1: Company Performance - 澜起科技's stock price increased by 7.53% to 185.7 HKD, with a trading volume of 5.29 billion HKD [1] - Citigroup's report indicates that the demand for CPU-based servers and new developments in AI applications could catalyze increased usage of DIMM memory modules, providing upward potential for the company's memory interface business [1] Group 2: Industry Trends - The arrival of the AI era is leading to rapid performance growth in computing chips, with interconnectivity becoming a bottleneck for AI computing clusters [1] - The company's interconnect chip business is expected to experience significant growth, driven by upgrades in memory interface technology from DDR4 to DDR5 and DDR6, as well as enhancements in PCIe from 4.0 to 6.0 and 7.0 [1] - The introduction of the CXL standard for memory pooling is anticipated to greatly increase the value of related interface chips [1]

Summary of Conference Call on Changjiang Electronics and Rambus Industry and Company Involved - The conference call primarily discusses **Changjiang Electronics** and **Rambus**, focusing on the semiconductor and memory interface industry trends. Core Insights and Arguments - **Rambus Performance Impact**: Rambus faced a decline in Q1 performance guidance due to quality issues in packaging and testing, leading to a significant drop in stock price. However, this does not alter the overall positive industry trend [1] - **Future Growth in Server Memory**: The company remains optimistic about future growth in server CPU memory modules and the ramp-up of MRDIMM products, indicating a sustained upward trend in the industry [1] - **Google TPUv8p Memory Pooling**: Insights from the supply chain suggest that Google's TPUv8p may introduce memory pooling to expand memory capacity, potentially adding hundreds of GBs of additional memory per TPU. This memory pooling is expected to gradually commercialize and scale up [1] - **CXL Expansion Players**: The main players in the CXL expansion chip market are identified as Changjiang Electronics and Rambus, with high-priced, high-margin products likely to see market scale expansion and profit growth in the context of industry ramp-up [2] - **AI Server Growth**: The AI server market is projected to experience significant growth, with the general server market also expected to show double-digit growth by 2025. This growth in AI and general server CPUs is anticipated to accelerate the memory module market [2] - **MRDIMM and Memory Interface Chips**: As MRDIMM scales up, the memory interface chips MRCD and MDB, which have higher price points, are expected to see significant volume growth [3] Other Important but Potentially Overlooked Content - **Investment Recommendations**: The call recommends investing in Changjiang Electronics, highlighting the recent release of AEC Retimer and the anticipated launch of switch chips. The long-term profit potential is estimated to reach 100-150 times earnings, with a target market capitalization of 300-450 billion [3] - **Rambus as a US Market Player**: Rambus is also mentioned as a key player in the US market, indicating its relevance in the broader semiconductor landscape [3]

Summary of Conference Call Records Companies and Industries Involved - **Companies**: Haiguang Information, Lianqi Technology, Wangsu Technology - **Industries**: AI computing, CDN (Content Delivery Network), semiconductor technology Key Points and Arguments Haiguang Information - Haiguang Information's market capitalization increased by over 90 billion RMB, leading the A-share market in January 2026 [2] - The company’s Deep Computing 3 has entered mass production, supporting FP8/FP4 precision, while Deep Computing 4 is expected to double performance, potentially becoming the strongest AI chip in China [3][7] - The estimated valuation for Haiguang's CPU is 900 billion RMB and for its GPU is 1.3 trillion RMB [3][7] - The company is projected to reach a market capitalization of over 2 trillion RMB by 2028, with a target of 1.2 trillion RMB for 2026 [8] Lianqi Technology - Lianqi Technology benefits from the growth in AI inference and supernode industries, particularly in memory interconnect chips, PCIe Retimer Switch, and CXL chips [1][2] - The company has made significant progress in the CXL field, with its products expected to be adopted by Google's next-generation TPU, creating a substantial incremental market [10] - Lianqi's revenue breakdown includes 90% from memory interconnect, 5% from PCIe CXL, and 5% from CPU and server-related products [10] Wangsu Technology - Wangsu Technology is the largest third-party neutral CDN company in China, with CDN business accounting for 60-70% of its revenue [11] - The company is benefiting from a near doubling of CDN prices in North America due to Google Cloud's price increase, indicating a reversal in the CDN and cloud computing price war [1][2][12] - Wangsu is expected to achieve a net profit of 1 billion RMB in 2026, with significant profit elasticity due to price increases, suggesting over 50% growth potential in its valuation [12] Capital Expenditure Trends - North America's top five CSPs are projected to have capital expenditures nearing 700 billion USD in 2026, a 50% increase from 400 billion USD in 2025, driven by Meta and Microsoft's unexpected capital spending [4] - Domestic internet capital expenditure in China is expected to reach 570-600 billion RMB in 2026, with growth anticipated to surpass that of overseas markets by 2027 due to advancements in self-developed chips and easing of restrictions [4] AI Inference Demand - The emergence of applications like MudBot is driving exponential growth in data and computing power consumption, shifting traffic from human-driven to robot-driven, enabling 24/7 usage [5] Supply-Side Technological Advances - Future server architectures are expected to adopt supernode technology, which will enhance cluster efficiency through memory pooling and high-speed interconnects [6] Other Notable Companies - Additional companies to watch include DingTong Technology, Zhongke Shuguang, Shuguang Shuchuang, Feirongda, Yingweike, and application vendors like Shuiyou Co. and Keda Xunfei, all of which show promising development prospects [13]

Group 1 - Wells Fargo indicates that Marvell Technology (MRVL.US) plans to acquire XConn for $540 million, which is deemed crucial for memory pooling and expected to enhance company earnings in the near term [1] - Analyst Aaron Rakes emphasizes that the acquisition further validates the importance of memory pooling technology in high-performance and competitive hardware solutions, particularly in supporting larger models and improving inference performance [1] - The proposed transaction, with 60% cash and 40% stock payment, is expected to contribute to revenue starting from the second half of the current fiscal year, potentially reaching $100 million by fiscal year 2028 [1] Group 2 - In the AI 2.0 era, the core contradiction in computing power development has shifted from merely "not fast enough" to "data handling cannot keep up" [2] - The emergence of CXL (Compute Express Link) technology represents a significant transformation of traditional computing models at the physical architecture level, enhancing AI computing power through memory decoupling, capacity expansion, and communication collaboration [2] - CXL technology is not just about increasing bandwidth; it reconstructs fragmented data centers into a cohesive working whole through resource pooling, capacity decoupling, and consistent communication, serving as a foundational technology for the transition from "single performance competition" to "cluster efficiency game" in AI computing [2]

Core Viewpoint - The article emphasizes the significance of the latest CXL 4.0 specification in enhancing memory connectivity and performance for high-performance computing, particularly in artificial intelligence applications [2][13]. Group 1: CXL 4.0 Specification Features - CXL 4.0 doubles the bandwidth to 128GTs without additional latency, enhancing data transfer speeds between connected devices [4][11]. - It supports high-speed data transfer between CXL devices, improving overall system performance [7]. - The specification retains full backward compatibility with CXL 3.x, 2.0, 1.1, and 1.0 versions, ensuring a smoother transition for existing deployments [12]. Group 2: Importance of CXL for AI - CXL addresses memory bottlenecks in AI workloads by enabling memory pooling, allowing all processors to access a unified shared memory space, thus improving memory utilization [15][17]. - It facilitates large-scale inference by providing quick access to large datasets without the need for memory duplication across GPUs [18]. - CXL is designed to meet the growing performance and scalability demands of modern workloads, particularly in AI and high-performance computing [19]. Group 3: Future Implications of CXL - The introduction of CXL is seen as a fundamental shift from static, isolated architectures to flexible, network-based computing, paving the way for next-generation AI and data-intensive systems [20]. - CXL enables a unified, flexible AI architecture across server racks, crucial for training large language models efficiently [21]. - Major industry players, including Intel, AMD, and Samsung, are beginning to pilot CXL deployments, indicating its growing importance in the semiconductor landscape [21].

Core Insights - Intel and Alibaba Cloud announced a series of cloud instances and storage solutions based on the new generation Xeon® 6 processors, addressing the challenges posed by AI scalability on cloud infrastructure [1][9] - High performance, high elasticity, and low total cost of ownership (TCO) are becoming key competitive indicators for global cloud providers [1] Group 1: Cloud Infrastructure Innovations - The introduction of "memory pooling" and flexible architecture is transforming cloud infrastructure, allowing dynamic allocation of resources based on demand [2][6] - Alibaba Cloud has deployed a unified hardware architecture across 29 global regions and 91 availability zones, enabling rapid resource allocation in response to sudden computing demands [4][9] Group 2: AI and Heterogeneous Computing - AI-driven heterogeneous computing is redefining the role of CPUs as central coordinators, with Intel integrating AMX matrix acceleration instruction sets to support various precision calculations [7] - The Xeon® 6 processors can efficiently handle large AI models, demonstrating significant performance improvements in various applications, such as data preprocessing for autonomous driving [7][8] Group 3: Collaboration and Competitive Edge - The stability and engineering support of the collaboration between Intel and Alibaba Cloud are highlighted as foundational elements for their long-term partnership [8] - The optimization of both hardware and software is becoming a key differentiator in the market, with Alibaba Cloud leveraging CXL 2.0 memory pooling technology for enhanced performance [8][9] Group 4: Future Directions - The shift from cloud adoption to intelligent cloud solutions is seen as an inevitable development path, with AI moving into a phase of large-scale application [9][10] - The collaboration between Intel and Alibaba Cloud aims to provide scalable and sustainable pathways for various industries through enhanced hardware performance and optimized software stacks [9][10]

Core Viewpoint - The launch of the CXL3.1 memory expansion controller (M88MX6852) by 澜起科技 marks a significant advancement in memory architecture, aimed at enhancing bandwidth and reducing latency for next-generation data center servers [1][2]. Group 1: Product Features - The M88MX6852 chip supports CXL.mem and CXL.io protocols, providing high bandwidth and low latency memory expansion and pooling solutions [1]. - It utilizes a PCIe 6.2 physical layer interface with a maximum transmission rate of 64 GT/s (x8 channels) and features dual-channel DDR5 memory controller supporting speeds up to 8000 MT/s [1]. - The chip integrates dual RISC-V microprocessors for dynamic resource configuration and hardware-level security management, along with multiple interfaces for system integration [1]. Group 2: Market Demand and Applications - The demand for cloud computing resource pooling is increasing, making traditional memory architectures a performance bottleneck [2]. - The CXL3.1 memory expansion controller enables elastic allocation and efficient utilization of memory resources, thereby reducing total cost of ownership (TCO) [2]. - The chip is compatible with EDSFF (E3.S) and PCIe add-in card (AIC) formats, making it suitable for various deployment environments including servers and edge computing [2]. Group 3: Industry Feedback - Stephen Tai, the company president, highlighted that the chip represents a breakthrough in CXL technology, enhancing memory expansion performance and energy efficiency [2]. - Feedback from industry leaders like Samsung and AMD indicates strong support for the CXL3.1 controller, emphasizing its role in advancing memory decoupling architecture and reducing TCO in data centers [2][3].