中国信息通信研究院首席专家石友康出席研讨会并致辞，中国信息通信研究院云大所总工程师郭亮主 持。中国移动（600941）云能力中心项目总师周宇，华为数据存储产品线战略与业务发展部总裁王旭 东，北京硅基流动科技有限公司解决方案总监唐安波发表主题演讲。 研讨会上，中国信息通信研究院首席专家石友康深刻阐述了当前AI规模化应用推进，推理环节的成 本、效率、质量问题凸显，先进存力成为提升AI推理效能、控制成本的关键。当前，国家高度重视先 进存力发展，在《算力基础设施高质量发展行动计划》等政策中明确提出"加速存力技术研发应用""持 续提升存储产业能力""推动存算网协同发展"，为产业发展指明了方向。中国信息通信研究院在政策研 究、标准制定、测试服务等方面开展多项工作，并联合产业链企业成立"算力产业发展方阵先进存力AI 推理工作组"，同时提出了三点建议：鼓励前沿存储技术研发创新，推动存算运深度融合，加强存算协 同产业生态建设，呼吁业界同仁凝聚共识，共同推动我国存算协同发展。 唐安波在会上围绕大模型推理"推不动、推得慢、推得贵"问题展开分享，硅基流动构建的AI infra工具 链，聚焦提升算力利用率。核心推理框架适配100多款开 ...

Core Insights - The AI industry is undergoing a structural shift, moving from a focus on GPU scaling to the importance of storage capabilities in enhancing AI performance and cost efficiency [1][10] - The demand for advanced storage solutions is expected to rise due to the increasing requirements of AI applications, with storage prices projected to remain bullish through Q4 2025 [1][10] - The transition from a "parameter scale" arms race to a "inference efficiency" commercial competition is anticipated to begin in 2025, emphasizing the significance of token usage in AI inference [2][10] Storage and Inference Changes - The fundamental changes in inference loads are driven by three main factors: the exponential growth of KVCache capacity due to longer contexts, the complexity of multi-modal data requiring advanced I/O capabilities, and the need for consistent performance under high-load conditions [4][10] - The bottleneck in inference systems is increasingly related to storage capabilities rather than GPU power, as GPUs often wait for data rather than being unable to compute [5][10] - Enhancing GPU utilization by 20% can lead to a 15%-18% reduction in overall costs, highlighting the importance of efficient data supply over merely increasing GPU numbers [5][10] New Storage Paradigms - Storage is evolving from a passive role to an active component in AI inference, focusing on data flow management rather than just capacity [6][10] - The traditional storage architecture struggles to meet the demands of high throughput, low latency, and heterogeneous data integration, which hinders AI application deployment [7][10] - New technologies, such as CXL and multi-level caching, are being developed to optimize data flow and enhance the efficiency of AI inference systems [6][10] Future Directions - The next three years will see a consensus on four key directions: the scarcity of resources will shift from GPUs to the ability to efficiently supply data to GPUs, the management of data will become central to AI systems, real-time storage capabilities will become essential, and CXL architecture will redefine the boundaries between memory and storage [10][11][12] - The competition in AI will extend beyond model performance to the underlying infrastructure, emphasizing the need for effective data management and flow [12]

次数）不足，导致GPU等昂贵算力资源长时间空闲。传统存储架构难以兼顾高吞吐、低时延及异构数据 融合的需求，造成业务发展瓶颈，阻碍AI应用落地。华为针对AI推理研发的UCM推理记忆数据管理技 术在行业落地中的核心作用，通过"集中高质数据、提速AI训练、优化推理效能"三个角度，打造AI推理 加速解决方案。 唐安波在会上围绕大模型推理"推不动、推得慢、推得贵"问题展开分享，硅基流动构建的AI infra工具 链，聚焦提升算力利用率。核心推理框架适配100多款开源大模型，并通过公有云服务平台为广大开发 者和企业提供优质的大模型服务。解决方案上，结合UCM技术卸载KVCache释放显存、提升推理性 能，同时通过智能网关进一步优化流量调度、弹性扩缩容等策略，来应对高并发、低延时、高吞吐、长 上下文等痛点，基于存储的KVCache方案可大幅提升系统吞吐。 在内部研讨环节，参会专家围绕算力产业发展方阵先进存力AI推理工作组下一阶段工作建议进行研 讨，中国移动云能力中心、华为、硅基流动、浪潮信息、清微智能、东方算芯、智元芯、算苗科技、得 一微电子等工作组成员单位参与讨论，贡献实践经验。存力中国行暨先进存力AI推理工作研讨会的成 ...

2023年10月，工信部等六部门印发《算力基础设施高质量发展行动计划》，明确提出"加速存力技术研 发应用"、"持续提升存储产业能力"、"推动存算网协同发展"等发展方向。 中国信通院联合产业链企业成立"算力产业发展方阵先进存力AI推理工作组"。石友康建议，鼓励前沿存 储技术研发创新，推动存算运深度融合，加强存算协同产业生态建设，业界凝聚共识，共同推动我国存 算协同发展。 华为数据存储产品线战略与业务发展部总裁王旭东表示，AI时代，IT基础设施能力面临三大挑战："管 不好"的数据、"喂不饱"的算力、"降不下"的成本。华为针对AI推理研发的UCM推理记忆数据管理技 术，通过"集中高质数据、提速AI训练、优化推理效能"三个角度，打造AI推理加速解决方案。 中新网北京11月5日电 (记者刘育英)由中国信息通信研究院组织召开的"存力中国行"北京站活动4日在北 京举行，活动通过座谈会和实地调研的方式，深入探讨AI时代下先进存力赋能AI大模型发展的支撑作 用。 当前，AI推理在各行各业大规模使用，各行各业迫切需要将模型能力无缝融入到实际业务场景中，如 投研分析、卷宗分析、智能客服、医疗影像辅助诊断等。AI技术虽已在文档处理等 ...