由英特尔资深人士创立的芯片初创公司Delos Data为人工智能网络筹集了1亿美元资金

路透中文
Sep 15
由<a href="https://laohu8.com/S/INTC">英特尔</a>资深人士创立的芯片初创公司Delos Data为人工<a href="https://laohu8.com/S/5RE.SI">智能</a>网络筹集了1亿美元资金

Stephen Nellis

路透旧金山9月15日 - 由英特尔公司(INTC.O)资深人士创立的初创公司Delos Data周二表示,已筹集1亿美元资金,用于开发芯片和软件,以加快数据在人工智能数据中心内部的传输速度,这些数据中心正以前所未有的速度变得更加复杂。

在人工智能热潮的早期阶段,数据中心通常主要由英伟达的图形处理单元(GPU)组成,这些GPU通过英伟达专有的网络技术 (link) 相互连接。但随着这些数据中心的重点从训练人工智能系统转向处理用户请求——即让人工智能“代理”代为执行任务——英伟达的竞争对手,如Cerebras Systems CBRS.O 和超微半导体(AMDAMD.O,已 (link) (link) 获得市场认可。

就连英伟达自身也已开始提供多种 (link) 计算 (link) 芯片。针对这一新兴的计算芯片多样化趋势,Delos Data设计了一套复杂的网络芯片和软件组合,其目标非常简单:帮助数据中心所有者让所有昂贵的计算设备以最快的速度进行通信,并具备足够的灵活性,以适应数据中心所有者当前或未来可能选择的任何计算芯片组合。

“我们不知道面向自主智能 $(AI)$ 的下一代基础设施架构会是什么样,”Delos Data 首席技术官兼联合创始人丹·戴利(Dan Daly)在接受采访时表示。“但我们确信,我们能够提供传输数据的最快捷、最高效的方式。”

英特尔前首席执行官帕特·格尔辛格(Pat Gelsinger)现为风险投资公司Playground的合伙人,该公司参与了Delos Data的本轮融资。他表示,计算芯片因等待数据而闲置,仍是人工智能数据中心中资金和能源的最大浪费之一。

“即使我拥有全世界的所有计算机,但如果无法让这些设备之间有效通信,那我拥有的只是一堆发热的硬件,它们在消耗电能,却只能闲置无事,”格尔辛格说道。

本轮融资的其他投资者包括Matrix Partners、Playground、Socratic Partners、Capricorn、Matter Venture Partners、IAG和DYNAMIQ。

At the request of the copyright holder, you need to log in to view this content

Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.

Most Discussed

  1. 1
     
     
     
     
  2. 2
     
     
     
     
  3. 3
     
     
     
     
  4. 4
     
     
     
     
  5. 5
     
     
     
     
  6. 6
     
     
     
     
  7. 7
     
     
     
     
  8. 8
     
     
     
     
  9. 9
     
     
     
     
  10. 10