AIDC解决方案
Search documents
以电强算以冷强算 华为 DCCP 峰会擘画 AIDC 生态合作新蓝图
Zhong Guo Qi Che Bao Wang· 2026-02-04 09:46
近日,以 "谊切苔岑,共赴山海" 为主题的 2026 年度 DCCP 卓越伙伴峰会圆满举办。本 次峰会汇聚数据中心能源及关键供电产业的全球精英伙伴,以深化 "伙伴 + 华为" 合作体系 为核心,集中传递华为数字能源业务与产品发展战略,全面沟通伙伴合作政策,深度倾听伙 伴发展诉求,携手全球伙伴抢抓产业机遇、实现互利共赢,共同迈向更大的商业成功。 锚定 AIDC 黄金时代 升级 "伙伴 + 华为" 合作体系 华为数字能源副总裁何波在开场致辞中指出,AI 驱动的第四次工业革命已全面开启, 万亿级 AIDC 基建成为全球最确定的核心投资方向。在产业机遇背后,AI 数据中心建设也 面临着电力缺口、高密供电、高密散热及交付运维等多重挑战。对此,华为坚持以电强算、 以冷强算的核心思路,通过四大核心链路的持续技术创新、建设模式的重构升级,打造安全 可靠、弹性敏捷、绿色低碳的 AIDC 解决方案,实现 "电进得去、热出得来",让每一瓦特 产出更多 Token,为 AI 世界的稳定运行筑牢根基。 何波强调,产业的繁荣绝非独行可至,唯有众行方能致远。面向未来,华为将持续升级 全球伙伴合作政策,向伙伴开放更多技术平台与优质资源,加大 ...
人工智能数据中心驱动储能产业变革
中国能源报· 2026-01-10 00:40
Core Viewpoint - The explosive growth of Artificial Intelligence Data Centers (AIDC) is reshaping the energy storage industry from three dimensions: market demand, application scenarios, and power system architecture [1][3][5]. Group 1: Market Demand - Global AIDC computing power demand is expected to grow exponentially, with data center electricity consumption projected to increase by 4 to 5 times by 2030. In China and Europe, green data centers will dominate [5]. - By 2030, the AIDC sector alone is anticipated to generate over 300 GWh of energy storage installation demand, becoming a new growth pole for the energy storage industry [5]. Group 2: Application Scenarios - AIDC's characteristics, such as high power density and significant load fluctuations, necessitate a shift in energy storage systems from "backup power" to "core power regulation" [5]. - The layout of AIDC is driving the evolution of energy storage application scenarios from centralized to a "centralized + distributed" collaborative model, requiring deep integration with energy storage systems [6]. Group 3: Power System Architecture - The central energy router is a product of structural changes, converting medium voltage grid electricity to 800V DC, adapting to AIDC's higher voltage and efficiency trends while simplifying the power supply chain [6]. - AIDC's energy demand is evolving from simple power supply assurance to the need for digital and intelligent energy scheduling, prompting a transition in the energy storage industry towards composite products that combine capacity and power [5]. Group 4: Challenges - AIDC's stable operation under extreme conditions is a primary challenge, as many AIDC facilities are located in remote areas with harsh environmental conditions [8]. - High reliability is crucial for AIDC energy storage, with a requirement for power supply online rates exceeding 99.9%. Any power interruption could lead to significant economic losses [8]. - Cost control is essential for the marketization of AIDC energy storage, as high investment and operational costs directly impact operational efficiency [9]. Group 5: Technology Iteration - The technology iteration direction for energy storage is becoming clearer, focusing on long-duration storage, high voltage architecture, multi-technology integration, and intelligent scheduling [11]. - Long-duration storage is key to addressing the structural contradiction between rapid computing power iteration and long power construction cycles [11]. - High voltage DC power supply architecture is becoming mainstream, with products like the central energy router simplifying supply chains and reducing energy loss [11]. - The integration of lithium and sodium battery technologies is being explored to meet AIDC's diverse needs, with each technology complementing the other [12]. - Intelligent scheduling and digital management are seen as core competitive advantages, leveraging big data and AI for real-time load monitoring and predictive adjustments [12].
SuperX与华胜天成成立合资公司 完善AI基础设施全生命周期解决方案
Zheng Quan Shi Bao Wang· 2025-10-31 08:51
Core Insights - SuperX and Hua Sheng Tian Cheng have established a joint venture, SuperX Global Service Pte. Ltd., in Singapore to provide end-to-end professional services for SuperX's global AI factory project [1][2] - The joint venture aims to address the increasing global demand for computing power, which traditional data centers struggle to meet due to the high-density computing, fast interconnectivity, and efficient cooling requirements of AI workloads [1] - SuperX's modular AI factory solution simplifies the construction of AI data centers (AIDC) into a standardized process, significantly reducing construction time and deployment difficulty [1] Company Collaboration - Hua Sheng Tian Cheng has extensive experience in providing digital services to over 16,000 clients, managing large-scale complex projects, and operating a global delivery network [2] - The joint venture will offer four core services: global contact center, implementation and installation services, after-sales maintenance services, and operational management services [2] - The establishment of the joint venture is expected to integrate Hua Sheng Tian Cheng's global service network with SuperX's product technology, shortening sales cycles and accelerating the deployment of complex solutions across various industries [2] Value Chain Enhancement - Previously, SuperX's core value was primarily in the sales of advanced hardware products and solutions; the joint venture fills a critical gap in SuperX's value chain by providing full lifecycle services from deployment to long-term maintenance and optimization [2] - This development is anticipated to significantly enhance customer loyalty, transforming one-time equipment transactions into long-term service and value co-creation partnerships [2] - SuperX is expected to gain trust from large enterprise clients more quickly, facilitating the global implementation of its modular AI factory solutions [2]
10倍能耗增长!阿里、抖音开启储能新需求
行家说储能· 2025-09-24 12:25
Core Insights - The article discusses the rapid growth of energy storage in data centers, driven by major tech companies like Alibaba and Douyin, as they ramp up AI infrastructure and energy consumption [2][3][6]. Group 1: Alibaba's Energy Storage Initiatives - Alibaba is investing 380 billion in AI infrastructure, expecting data center energy consumption to increase tenfold by 2032 [3]. - To manage this increased energy demand, Alibaba aims to reduce its Power Usage Effectiveness (PUE) from 1.25 to below 1.1, focusing on technologies like immersion cooling, photovoltaic storage systems, and photonic chips [3]. - A joint project with Mingyang Smart Energy will launch in 2024, featuring a 200 MW wind power and a 40 MW/160 MWh energy storage system, with an expected annual output of 797 million kWh [3][4]. Group 2: Douyin's Energy Storage Plans - Douyin plans to procure over 200 MWh of large-scale lithium battery storage and associated wind and solar equipment, with a project launch set for November 2025 [6]. - The energy consumption of Douyin's data centers is among the highest in the internet industry, with a single large data center projected to consume over 1 billion kWh annually in 2024 [7]. Group 3: Industry-Wide Trends - Other tech companies, including Tencent, are also investing in data center energy storage, with Tencent's Tianjin data center expected to produce 12 million kWh of green electricity annually [8]. - By 2030, global data center energy storage capacity is projected to reach 262.5 GWh, with China's market expected to exceed 100 GWh [8]. - Companies like CATL, BYD, and others are positioning themselves to capture market share in the rapidly growing energy storage sector, with a projected domestic market size of 800 GWh over the next three years [10].