Moor Thread, in collaboration with China Mobile Research Institute and others, jointly releases the 128-card high-density super node reference design

robot
Abstract generation in progress

Recently, leveraging the OISA Collaborative Innovation Platform, industry partners such as Moore Thread, China Mobile Research Institute, and Zhijiang Laboratory officially released the “OISA High-Density Super Node Reference Design Technical Specification.” This specification addresses current challenges faced by intelligent computing centers, including interconnection bottlenecks, power supply pressure, and heat dissipation limits, proposing a full-stack solution that lays a key technological foundation for building autonomous, controllable, high-performance intelligent computing clusters. In the physical layout and logical topology of intelligent computing centers, the OISA high-density super node reconstructs spatial density and expansion dimensions deeply, breaking through traditional architecture barriers to computing power growth. The technical specification boldly introduces large-dimension high-density cable solutions, achieving full interconnection of 128 cards within a standard single-width cabinet based on mainstream 32- to 64-card interconnects, and supporting expansion to 256 cards through cabinet stacking, greatly optimizing computing power output per unit land area. (Moore Thread)

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