Market Overview - The A-share market saw all three major indices rise collectively, with the Shanghai Composite Index increasing by 1.36%, surpassing the 3900-point mark; the Shenzhen Component Index rose by 2.06%, and the ChiNext Index increased by 3.02% [1] - The total market turnover reached 1.8927 trillion yuan, an increase of 141.4 billion yuan compared to the previous trading day, with over 4600 stocks rising [1] Sector Performance - The computing power sector reignited market enthusiasm, particularly the CPO (Co-Packaged Optics) segment, which showed significant momentum [3][7] - Leading stock Yuanjie Technology hit the daily limit, achieving a historical high, while Industrial Fulian topped the net inflow list with 2.207 billion yuan, marking a 9.57% increase in its stock price [3] - Other notable performers included New Yisheng and Luxshare Precision, with net inflows of 1.792 billion yuan and 1.52 billion yuan, respectively [3] Banking Sector - The banking sector experienced a strong rebound, with Agricultural Bank of China rising by 1.68%, marking its 13th consecutive day of gains and reaching a historical high [5] Computing Power Industry - The computing power industry is witnessing breakthroughs, with Alibaba Cloud's Aegaeon system significantly reducing model switching costs by 97% and improving request processing capabilities by 2-2.5 times [10] - Citic Securities expressed optimism about the rapid iteration of domestic large models and the seamless adaptation of domestic computing power chips, which are expected to drive the continuous development of domestic AI [10] Optical Module Market - According to Guosheng Securities, the optical module market is experiencing rapid growth driven by the explosion of AI computing power demand, with price changes reflecting technological iteration, cost control, and product structure optimization [11] - Leading companies in the optical module sector are expected to maintain strong profitability and competitive advantages due to the global data center construction and upgrade wave [11] Power Sector - The power sector saw a surge, with Shanghai Electric and Hengsheng Energy hitting the daily limit, while Yunnan Energy Investment and Hunan Development also saw gains [13] - According to CITIC Construction Investment, the global demand for electrical equipment is on the rise, driven by increased electricity demand from AI, leading to significant growth in supporting electrical equipment demand [14]

牛市来了还没上车？上天天基金APP搜索777注册即可领500元券包，优选基金10元起投！限 量发放！先到先得！ 10月21日，A股三大指数集体上涨。截至收盘， 上证指数涨1.36%，重回3900点上方；深证成指 涨2.06%、创业板指涨3.02%。 全市场全天成交额18927亿元，较上个交易日增加1414亿元。全 市场超4600只个股上涨。 消息面上，算力产业链近期亦有突破。阿里云微信公众号10月16日消息， 阿里云自研的多模型混 合服务系统Aegaeon成功入选国际顶级系统会议SOSP（操作系统原理研讨会）。该系统可通过精 确的执行时间预测和创新的调度算法规划是否需要切换模型，从而实现多个模型混合服务且满足延 迟要求。 从盘面上看，算力概念再次点燃市场热情。其中，CPO（共封装光学）板块持续发力，热点颇多。 截至收盘，龙头股源杰科技"20CM"涨停，股价创历史新高； A股主力资金净流入榜单中，工业富 联位列第一名， 净流入金额为22.07亿元，总市值1.35万亿元，该股涨9.57%。 新易盛、立讯精 密分列第二、第三名，净流入金额分别为17.92亿元、15.2亿元；中际旭创成交额突破244亿元， 位列A股第 ...

国盛证券研报称，在AI算力需求持续爆发的驱动下，光模块市场正经历着高速增长与技术迭代的双重变奏。光模块价格的变动反映了该行业健康发展的 本质——价格变动不是简单的供求博弈，而是技术迭代速度、成本控制能力与产品结构优化三者共同驱动。技术领先、客户优质、产能全球化的光模块龙 头企业将保持强劲的盈利能力和竞争优势，持续受益于全球数据中心建设与升级浪潮。综上，继续看好算力板块，推荐算力产业链相关企业如光模块行业 龙头。 10月21日，A股三大指数集体上涨。截至收盘，上证指数涨1.36%，重回3900点上方；深证成指涨2.06%、创业板指涨3.02%。全市场全天成交额18927亿 元，较上个交易日增加1414亿元。全市场超4600只个股上涨。 从盘面上看，算力概念再次点燃市场热情。其中，CPO（共封装光学）板块持续发力，热点颇多。 截至收盘，龙头股源杰科技"20CM"涨停，股价创历史新高；同花顺A股主力资金净流入榜单中，工业富联位列第一名，净流入金额为22.07亿元，总市值 1.35万亿元，该股涨9.57%。新易盛、立讯精密分列第二、第三名，净流入金额分别为17.92亿元、15.2亿元；中际旭创成交额突破244亿元，位 ...

Core Viewpoint - Alibaba Cloud's Aegaeon solution addresses the common issue of GPU resource waste in AI model services, significantly improving GPU utilization rates and has been recognized at the prestigious SOSP 2025 conference [2][4]. Group 1: Aegaeon Solution - Aegaeon was successfully selected for the SOSP 2025 conference, highlighting its innovative approach to solving GPU resource waste in AI model services [2][4]. - During a beta test lasting over three months, Aegaeon reduced the number of NVIDIA H20 GPUs required for serving large models from 1192 to 213, achieving an 82% reduction in GPU usage [5]. - The system's ability to pool GPU resources breaks the inefficient model-to-GPU binding, allowing for more effective resource allocation [8]. Group 2: Technical Innovations - Aegaeon's core innovation is token-level scheduling, which dynamically decides whether to switch models after generating each token, enabling fine-grained management of resources [8]. - The system can support up to seven different models simultaneously on a single GPU, improving effective throughput by 1.5 to 9 times and achieving 2 to 2.5 times the request processing capability compared to existing solutions [9]. - Aegaeon reduces model switching overhead by 97% through various optimizations, ensuring real-time responsiveness for model switching [8]. Group 3: Industry Implications - The integration of system software and AI model technology is emerging as a new trend, with a focus on optimizing underlying systems to better support AI applications [9]. - The future of AI development will rely not only on hardware advancements but also on software innovations that maximize existing hardware potential [9].