--- 格式版本: 2 标题: "Akash Systems Bets on Diamond Tech to Crack AI’s Thermal Ceiling" 原文链接: "https://www.datacenterknowledge.com/data-center-chips/interview-akash-systems-bets-on-diamond-tech-to-crack-ai-s-thermal-ceiling" 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: "标题附近明确标注的日期，且与发现时间一致，是最可能的文章发布时间" 发布时间校准置信度: "1" 发布时间候选数量: 9 发布时间严格候选数量: 9 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: "2026-08-17T20:58:39+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: 30375 发现时间: "2026-08-17T20:11:14+08:00" 入库时间: "2026-08-17T12:59:23.248Z" 来源平台: "Data Center Knowledge 搜索" 搜索渠道: "source_template" 搜索词: "https://www.datacenterknowledge.com/search?q=AMD" 匹配关键词: - "AI" - "GPU" - "Liquid Cooling" - "Immersion" - "deployment" - "performance" 相关厂家: - "AMD" - "NVIDIA" - "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: 69 AI分档: "召回候选" AI质检状态: "不通过" AI打分理由: "文章聚焦芯片级钻石冷却技术，与AI基础设施散热相关但与超节点/机柜级系统直接关联有限；来源为专业媒体，含近期商业发布和部署信息，技术细节中等，整体有价值但未达高置信优质。" AI质检模型: "ali-deepseek-v4-flash" AI质检时间: "2026-08-17T20:59:37+08:00" AI主题相关性: 12 AI来源权威性: 10 AI新颖性: 14 AI技术细节: 12 AI商业部署信号: 13 AI完整性: 8 AI摘要: "Akash Systems 正以钻石基冷却技术进军AI数据中心散热，该技术已商用适配AMD Instinct MI350 GPU，据报道获3亿美元订单，并已部署于英伟达H200平台。" AI摘要模型: "ali-deepseek-v4-flash" AI摘要时间: "2026-09-07T03:23:11.871Z" 采集批次: "2026年8月17日19点06分47秒" 采集批次ID: "20260817-190647-603" 去重键: "https://www.datacenterknowledge.com/data-center-chips/interview-akash-systems-bets-on-diamond-tech-to-crack-ai-s-thermal-ceiling" --- An Informa TechTarget Publication NEWSLETTER SIGN-UP DATA CENTER CHIPS COOLING DATA CENTER HARDWARE INFRASTRUCTURE INTERVIEWS Interview: Akash Systems Bets on Diamond Tech to Crack AI’s Thermal Ceiling With AMD and Nvidia partnerships and early deployments underway, Akash Systems is pushing diamond-based cooling into the AI mainstream – targeting heat as the next constraint on data center scale. Shane Snider,Senior News Writer,Data Center Knowledge March 24, 2026 6 Min Read IMAGE: AKASH SYSTEMS Akash Systems is stepping into the AI infrastructure spotlight with a growing list of partnerships and early deployments, positioning its diamond-based cooling technology as a tool to break through one of the industry’s toughest constraints – heat. The company recently announced that its diamond cooling technology is now commercially available on AMD’s Instinct MI350 series GPUs, backed by a reported $300 million launch order. At the same time, Akash says its technology has already been deployed on Nvidia H200 platforms, with support for next-generation Blackwell systems planned for future deployment. The momentum comes as data center operators face mounting pressure from AI workloads that are driving sustained, high-density compute. Racks are pushing beyond 100 kW, with future deployments expected to climb even higher as inference workloads begin to dominate. Against that backdrop, Akash is positioning its technology as a new layer in the cooling stack – one that operates directly at the chip level and complements broader industry shifts toward liquid and immersion cooling. Related:‘GPUs Suck’: Former Intel CEO Slams Data Center Hardware Limitations Crucially, the approach could also extend the life of existing infrastructure. For operators unable to retrofit facilities for liquid cooling, chip-level thermal improvements offer a path to deploy next-generation GPUs without massive capital overhauls. Akash’s origins highlight the durability of the approach. The company’s diamond-based thermal technology was initially developed for space applications, working with NASA and DARPA to operate in extreme environments before being adapted for terrestrial data centers. In this Q&A, Pamit Surana, co-founder and chief commercial officer of Akash Systems, discusses the company’s growing partnerships, the role of diamond in AI infrastructure, and why thermal management is becoming primary design challenge. DCK: You’ve announced support for both AMD Instinct MI350 and Nvidia H200, with Blackwell ahead. What does that say about where Akash fits in the AI ecosystem? Pamit Surana: It reflects how broadly the thermal challenge is being felt across the ecosystem. Whether it’s AMD or Nvidia platforms, power densities are increasing rapidly. Our goal is to sit at that chip level and provide a solution that works across architectures. The fact that we’re engaging across multiple GPU platforms shows that this isn’t a niche issue – it’s an industry-wide constraint. DCK: AI data centers are pushing power density to new extremes. Where does Akash Systems fit into that shift? Related:AMD Fires Back at Nvidia with Helios AI System, Epyc CPUs PS: AI infrastructure is fundamentally changing the thermal profile of data centers. What we’re seeing now is sustained high-power operation at levels that traditional materials struggle to handle efficiently. Our focus is on removing heat at the source – directly at the chip level – using diamond, which has significantly higher thermal conductivity than conventional materials. DCK: You’re moving into commercialization. What’s driving demand right now? PS: The demand is coming from multiple directions. Hyperscalers are pushing density higher, chipmakers are increasing power at the silicon level, and operators are running into real power and cooling constraints. At the same time, there’s a need to deploy infrastructure faster. Solutions that can improve efficiency without requiring a full redesign are getting a lot of attention. DCK: You’ve positioned diamond-based cooling as a breakthrough. What’s the real-world delta versus copper or aluminum? PS: Diamond is the most thermally conductive material available – it’s roughly five times more conductive than copper, which is the current industry standard. That allows us to reduce GPU temperatures significantly. In our deployments, we’ve seen up to a 10°C reduction under sustained workloads, which translates directly into better performance and efficiency. Related:Europe’s Chips Act: A Brief Supply Chain Opinion DCK: How important is that temperature delta in real-world AI deployments? PS: It’s critical. Lower temperatures mean you can run at peak performance without throttling, and you can do that using less energy. That has a direct impact on both performance and operating cost. At scale, those gains compound quickly. DCK: Diamond still sounds expensive. What’s changed to make this viable at scale? PS: We’re engineering synthetic diamond specifically for thermal applications. We’re not using gemstone-grade material. That allows us to optimize for manufacturability and cost while still delivering the performance benefits. As we scale production, the economics continue to improve. DCK: You’ve described this as a new layer in the cooling stack. How does it fit with liquid and air cooling? PS: We’re complementary. Liquid cooling addresses heat at the system or rack level. We’re removing heat at the source, before it even reaches those systems. When you combine both approaches, you can significantly improve overall efficiency. DCK: There’s a large installed base of air-cooled data centers. How does Akash fit into that reality? PS: That’s a key opportunity. Not every facility can transition to liquid cooling easily. Our technology allows operators to upgrade to next-generation GPUs while staying within existing infrastructure constraints. It’s a way to extend the life and capability of those data centers without major capital investment. DCK: You mentioned earlier roots in space – how did that shape the technology? PS: Our work with NASA and DARPA required us to solve thermal challenges in some of the most extreme environments possible. That pushed us to develop highly efficient, durable solutions. Bringing that technology into data centers was a natural progression once we saw the scale of the opportunity. DCK: With power constraints becoming a global bottleneck, how much efficiency can this unlock? PS: Better thermal management directly translates into better energy efficiency. If you can keep chips cooler, you reduce the energy overhead required for cooling and improve overall system performance. That has a meaningful impact at the data center scale. DCK: If inference overtakes training, how does that reshape cooling requirements? PS: It shifts everything toward sustained efficiency. Instead of optimizing for peak bursts, you need systems that can run continuously at high utilization. That’s where consistent thermal performance becomes critical. DCK: Are we heading toward a world where thermal management becomes as critical as compute itself? PS: In many ways, we’re already there. Compute performance is increasingly constrained by how effectively you can manage heat. DCK: In one sentence, what’s the future of data center cooling in the AI era? PS: Cooling will move closer to the chip and become a fundamental enabler of next-generation compute. From Space Tech To AI Factories Akash’s progress highlights a broader shift in the industry’s approach to thermal management. What was once a facility-level concern is now moving into silicon, packaging, and materials science. As AI infrastructure scales, incremental gains at the chip level can unlock outsized improvements across the entire stack. Akash’s partnerships with AMD and Nvidia – and its ability to operate within existing air-cooled environments – position it uniquely in a market that is balancing rapid innovation with practical deployment constraints. If the company can scale manufacturing to meet hyperscale demand, diamond-based cooling could emerge as a critical component of next-generation AI infrastructure, particularly as operators seek to maximize performance within limited power resources. 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 DATA CENTER CONSTRUCTION New Data Center Developments: June 2026 by James Walker JUN 03, 2026 ENERGY & POWER SUPPLY Why Data Centers Are Turning to Behind-the-Meter Power by Jack Vaughan APR 15, 2026 COOLING Data Center Cooling Methods: Costs vs. Efficiency vs. Sustainability by Christopher Tozzi OCT 31, 2025 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 SPONSORED CONTENT The AI era is defined by pushing the frontiers of what’s possible AUG 17, 2026 DATA CENTER SITE SELECTION Land Banking Explained: A Novel Strategy for Data Center Expansion AUG 14, 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