在AI全链条中，存储技术正从幕后走向台前。周跃峰比喻称："AI推理如同人类思考，需要快速调用历史记忆提升效率。"华为推出的UCM推理记忆数据管 理器，通过KV Cache核心架构实现记忆数据多级缓存，结合信息浓缩与智能联想技术，动态优化推理过程中的数据调用路径。该方案在金融领域试点中， 将AI服务响应延迟降低30%，算力成本下降25%，验证了存储革新对突破推理效率瓶颈的实效性。 作为连续多年稳居国内存储市场首位、全球第二的供应商，华为正通过技术开放推动生态共建。周跃峰透露，企业已将AI工具链、推理框架等核心软件能 力开源，与产业伙伴共建技术底座。他特别强调，存储技术应上升为AI战略基础设施，通过政企学研协同创新，加速先进数据基础设施落地，为我国在AI 下半场竞争中构筑领先优势。 随着人工智能技术深度融入经济社会发展各领域，数据要素的汇聚、流通与利用效率正成为全球AI竞争的关键。据国家数据局最新统计，2024年我国数据 产量突破41.06ZB，同比增幅达25%，这一爆发式增长对数据基础设施建设提出了更高要求。在第十五届智慧城市与智能经济博览会上，华为数据存储产品 线总裁周跃峰提出，需通过分层布局构建先进数据基础 ...

Group 1 - The 2025 China Computing Power Conference was held in Datong, Shanxi, focusing on accelerating towards an intelligent world through advanced solutions in connectivity, computing, storage, and digital energy [1][6] - Huawei's Vice President Zhou Yuefeng emphasized the importance of data aggregation, circulation, and value release for cities, industries, and enterprises to transition from a data powerhouse to a data strong nation, seizing opportunities in the AI era [1][3] Group 2 - China has become a global data powerhouse with an annual data output exceeding 40ZB, but the effective data retention rate is only 2.8%, indicating a significant amount of data is discarded at the source [3][4] - The construction of advanced data infrastructure is crucial, with a focus on creating trusted data management centers and efficient data circulation to transform data resources into valuable assets [3][4] Group 3 - At the industry level, the establishment of high-quality industry-specific data repositories is essential for AI model effectiveness, encouraging leading enterprises to build collaborative data-sharing platforms [4][5] - For enterprises, the development of AI data lakes is necessary to facilitate collaboration among multiple intelligent agents, enhancing the precision and real-time knowledge of applications [5][6] Group 4 - Huawei proposed the RAS concept for AI data center construction, focusing on reliability, agility, and sustainability to address the challenges of increasing computing power demands [6] - The company aims to strengthen core competitiveness and industry ecology by showcasing collaborative development in computing, storage, and green energy supply [6]