--- 格式版本: 2 标题: "Nebius Touts $10B, 310 MW Data Center Plan" 原文链接: "https://www.datacenterknowledge.com/data-center-construction/nebius-s-10b-310mw-data-center-plan-signals-ai-infrastructure-shift" 发布日期: "2026-08-17" 发布时间校准状态: "found" 发布时间需复核: "否" 发布时间来源: "llm:scrape:strict_markdown_body" 发布时间证据: "AUG 17, 2026" 发布时间校准原因: "该日期为候选中最接近发现时间且位于标题附近的日期，符合文章发布惯例。" 发布时间校准置信度: "1" 发布时间候选数量: 6 发布时间严格候选数量: 6 发布时间原页读取状态: "" 发布时间未找到原因: "" 发布时间校准时间: "2026-08-18T12:48:51+08:00" 发布时间仲裁状态: "confirmed" 发布时间仲裁尝试次数: 1 发布时间仲裁耗时毫秒: 7684 发现时间: "2026-08-18T12:43:18+08:00" 入库时间: "2026-08-18T04:49:11.918Z" 来源平台: "Data Center Knowledge 搜索" 搜索渠道: "source_template" 搜索词: "https://www.datacenterknowledge.com/search?q=Vera%20Rubin" 匹配关键词: - "Vera Rubin" - "AI" - "GPU" - "Liquid Cooling" - "deployment" - "performance" 相关厂家: - "NVIDIA" - "Meta" - "Microsoft" - "AWS" - "Google" 相关专家: [] 内容类型: "网页" 抓取工具: "CDP Render" 清洗工具: "CDP Text + Defuddle/Readability 正文提取" 原始附件: [] AI优质: "否" AI打分: 59 AI分档: "召回候选" AI质检状态: "不通过" AI打分理由: "文章聚焦Nebius数据中心投资与Meta/Nvidia合作，属AI基础设施商业部署，但未涉及超节点或机柜级架构细节，仅提及Vera Rubin平台，技术含量低。" AI质检模型: "ali-deepseek-v4-flash" AI质检时间: "2026-08-18T12:49:18+08:00" AI主题相关性: 8 AI来源权威性: 10 AI新颖性: 15 AI技术细节: 5 AI商业部署信号: 12 AI完整性: 9 AI摘要: "荷兰neocloud提供商Nebius宣布在芬兰拉彭兰塔建设100亿美元、310MW的AI数据中心，预计2027年上线，采用Nvidia Vera Rubin平台。" AI摘要模型: "ali-deepseek-v4-flash" AI摘要时间: "2026-09-07T03:23:01.991Z" 采集批次: "2026年8月18日10点53分10秒" 采集批次ID: "20260818-105310-367" 去重键: "https://www.datacenterknowledge.com/data-center-construction/nebius-s-10b-310mw-data-center-plan-signals-ai-infrastructure-shift" --- An Informa TechTarget Publication NEWSLETTER SIGN-UP DATA CENTER CONSTRUCTION DATA CENTER SITE SELECTION INVESTING AI DATA CENTERS NEWS Nebius’ $10B, 310 MW Data Center Plan Signals AI Infrastructure Shift The company’s 310 MW project, along with agreements with Meta and Nvidia, underscores how AI compute is moving from on-demand cloud to reserved capacity. Shane Snider,Senior News Writer,Data Center Knowledge March 31, 2026 4 Min Read ALAMY Netherlands-based neocloud provider Nebius on Tuesday unveiled plans for a $10 billion, 310 MW AI data center in Lappeenranta, Finland, following securing a multibillion-dollar agreement with Meta, as hyperscalers lock in external AI capacity. Meta has committed up to $27 billion over five years to secure compute capacity from Nebius, including $12 billion in dedicated infrastructure and an option to purchase an additional $15 billion from future deployments, the company said. Earlier this month, Nvidia backed Nebius with a $2 billion investment. The Lappeenranta campus is slated to begin coming online in 2027 and is designed for high-density AI workloads. Nebius said it will be built around Nvidia’s next-generation Vera Rubin platform. “We have been building in Finland for many years and are pleased to be expanding our presence here,” Nebius CEO Arkady Volozh said in a statement. “Lappeenranta represents a significant addition to our global AI infrastructure build-out and will make a significant contribution to achieving our goals.” Related:Neoclouds vs. Hyperscalers: Will AI’s Specialized Clouds Prevail? Nebius said the 100-acre Lappeenranta campus will create up to 700 construction jobs and 100 permanent roles once operational. The announcement follows the recent expansion of Nebius’ first Finnish data center in Mäntsälä to 75 MW, completed earlier this year. The company plans further expansion in Finland as it scales globally, including a 240 MW AI facility under development in France. New Phase of AI Expansion? At 310 MW, the planned Finland site places Nebius among a growing group of operators deploying hyperscale infrastructure specifically for AI training and inference. The inclusion of a major hyperscale customer suggests that more AI infrastructure projects are being developed against committed demand rather than on a speculative basis. Such projects require access to advanced chips as well as substantial power, cooling capacity, and land – factors that are increasingly primary constraints in AI deployment. “The AI infrastructure bottleneck is power, permitting, and deployment speed, not compute demand,” Patrick Moorhead, CEO and chief analyst at Moor Insights & Strategy, told Data Center Knowledge. “Even Meta cannot build capacity fast enough internally.” Finland has emerged as an attractive location due to relatively abundant power resources and favorable climate conditions for cooling, aligning with broader industry efforts to optimize energy efficiency and site selection. Moorhead said the location choice reflects those dynamics, pointing to energy availability and climate advantages as key drivers for large-scale AI deployments. Related:Europe’s Data Center Market Enters a Pivotal Phase Amid Structural Challenges Neocloud Momentum Builds The Nebius buildout comes as a new class of AI-native infrastructure providers gains traction. “These neoclouds are purpose-built for AI from the ground up – from the racks and silicon to the networking,” said Matt Kimball, vice president and principal analyst at Moor Insights & Strategy. “That level of specialization is difficult to replicate in traditional cloud environments and is why they’re not going away anytime soon.” “Purpose-built AI cloud providers move faster because they have no legacy infrastructure,” Moorhead added. “That speed and power advantage is real today.” Often positioned between hyperscalers and smaller GPU cloud operators, neocloud providers focus on high-density deployments optimized for AI workloads and more flexible approaches to capacity provisioning. Kimball said enterprises are increasingly adopting a federated approach to AI infrastructure, distributing workloads across on-premises systems, public cloud platforms, and specialized providers. “Five years out, enterprises won’t rely on a single environment,” he said. “You’ll see AI workloads spread across on-premises, cloud, and neocloud infrastructure – in some cases without the enterprise even knowing where specific workloads are running.” Related:Hyperscaler Capex Snowballs Toward $700B as Firms Stage AI Builds He added that performance advantages extend beyond hardware. “It’s not just about chips – it’s the full stack,” Kimball said. “AI inference is only as fast as the data pipeline behind it, and these providers are optimizing across the entire system.” Kimball also pointed to Nvidia’s investment activity as validation of the model. “When you look at what Nvidia is doing, they’re investing in what are effectively AI token factories,” he said. “They recognize how critical these providers are to both near-term and long-term AI enablement.” Durability Questions Remain Despite strong near-term demand, questions remain about the long-term competitive position of neocloud providers. “I remain skeptical about long-term durability,” Moorhead said. “The hardware is all Nvidia, and while Nebius has built its own infrastructure, the key differentiator is the orchestration software layer – and that’s a relatively thin moat.” He added that enterprise adoption could be a limiting factor. “These providers have virtually no enterprise sales force,” Moorhead said. “To attract more enterprise customers, they’ll need to expand services and partner more closely with infrastructure vendors.” Moorhead said the competitive landscape could shift as hyperscalers expand their own capabilities. “AWS, Microsoft, and Google are scaling aggressively, including custom silicon alongside Nvidia,” he said. “They already have the enterprise relationships, security posture, and global infrastructure that these providers lack.” Still, he pointed to areas where neocloud providers may retain an edge. “Elastic GPU burst capacity and AI factory disaster recovery are real use cases,” Moorhead said. “Those are harder for hyperscalers to deliver cost-effectively.” He added that Nvidia’s backing underscores the strategic importance of the segment. “Nvidia’s investment signals that these AI clouds are critical distribution channels for GPUs in this phase of the market,” Moorhead said. About the Author Shane Snider Senior News Writer, Data Center Knowledge Shane Snider is Senior News Writer at Data Center Knowledge, covering AI infrastructure, hyperscale data centers, cloud platforms, and the power and energy systems driving modern compute expansion. His reporting focuses on the operational, economic, and environmental forces reshaping digital infrastructure, including AI factories, utility constraints, liquid cooling, renewable energy procurement, and next-generation data center architectures. He has won recent Azbee awards for news series and government reporting. Based in Raleigh, North Carolina, Snider covers how hyperscalers, utilities, chipmakers, and infrastructure providers are responding to the rapid rise of AI workloads and global compute demand. You can reach Shane at shane.snider@informatechtarget.com or on LinkedIn. Want more Data Center Knowledge stories in your Google search results? ADD US NOW Subscribe to the Data Center Knowledge Newsletter Get analysis and expert insight on the latest in data center business and technology delivered to your inbox daily. NEWSLETTER SIGN-UP Recommendations SPONSORED The AI era is defined by pushing the frontiers of what’s possible AUG 17, 2026 ENERGY & POWER SUPPLY Amazon Explores Grid Connection for 8,000-Acre AI Campus by Shane Snider AUG 17, 2026 ENERGY & POWER SUPPLY PJM’s New Deal for Data Centers: Bring Power or Face Cuts by Shane Snider AUG 17, 2026 NETWORKING Zayo, Nvidia Target Fiber Capacity as AI Infrastructure Spreads by Shane Snider AUG 13, 2026 Editor's Choice DATA CENTER SITE SELECTION Virginia Report Finds Groundwater Running Dry for New Data Centers JUL 30, 2026 PHYSICAL SECURITY Data Centers Under Fire: Do Facilities Need Counter-Drone Defenses? JUL 30, 2026 OUTAGES Fault in Data Center Alley Triggered 3 GW Load Drop JUL 23, 2026 ENERGY & POWER SUPPLY DOE: AI Data Centers Are Transforming America’s Transmission Map JUL 24, 2026 SUSTAINABILITY After the AI Rush, Can Data Centers Reclaim Sustainability? JUL 23, 2026 Want more Data Center Knowledge stories in your Google search results? Subscribe to the Data Center Knowledge Newsletter Get analysis and expert insight on the latest in data center business and technology delivered to your inbox daily. NEWSLETTER SIGN-UP Industry Voices ENERGY & POWER SUPPLY Amazon Explores Grid Connection for 8,000-Acre AI Campus AUG 17, 2026 ENERGY & POWER SUPPLY Sunrun, Voltus Bring Home Batteries Into AI Capacity Push AUG 17, 2026 DATA CENTER WORLD POWER Join the entire data center ecosystem September 21 – 23, 2026, at the Gaylord Texan Resort & Convention Center in Dallas, TX.  LEARN MORE Discover More InformationWeek Data Center World AFCOM Working With Us About Us Meet Our Writers Reprints Advertise Contact Us Join Us NEWSLETTER SIGN-UP Follow Us Copyright © 2026 TechTarget, Inc. d/b/a Informa TechTarget. This website is owned and operated by Informa TechTarget, part of a global network that informs, influences and connects the world’s technology buyers and sellers. All copyright resides with them. Informa PLC’s registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. TechTarget, Inc.’s registered office is 275 Grove St. Newton, MA 02466. Home| CCPA: Do not sell my personal info| Cookies| Privacy| Terms Your Privacy Choices