SemiAnalysis: AI networks are not a choice between copper cables and optical communication; both will benefit from the expansion of GPU clusters

robot
Abstract generation in progress
Mars Finance News reported that on June 20, SemiAnalysis, an independent research institute focused on semiconductors and AI, said that investors are increasingly viewing AI networks as a binary choice between copper cabling and optical communication, while continually assessing the rotation between the two types of themes. However, it believes that as the size of GPU clusters grows, the importance of efficiently connecting these clusters will only continue to rise, driving growth in demand for network content of all kinds. Copper cabling and optical communication play different but complementary roles. When copper cabling can meet requirements for distance, power consumption, cost, and reliability, copper cabling remains the preferred solution; while when bandwidth and distance exceed the copper cable’s actual limitations, optical communication becomes necessary. NVIDIA’s strategy has remained consistent: use copper cabling when copper cabling can be used, and adopt optical communication when optical communication is required. This is not a market of either/or. Growth in optical interconnects does not mean copper cabling will disappear, and resilient demand for copper cabling does not mean optical communication will not grow. As AI systems increase in scale and complexity, the importance of data transmission between GPUs, switches, racks, and clusters will rise, and both copper and optical interconnects are expected to benefit.
View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pinned