Silicon Photonics Cold Plate Liquid-Cooled Switch
8Tbps of bandwidth using 64 electrical lanes and incorporates an integrated liquid-cooled cold plate capable of supporting 400W+ module power consumption. Graphics processing unit (GPU) computing clusters, which serve as the basic architecture to support AI, ML, and similar applications, raise higher requirements for network transmission than central processing unit (CPU) common computing clusters. While the industry-standard OSFP (Octal Small Form-Factor Pluggable) module has successfully enabled 400Gbps, 800Gbps, and 1. 8 Tbps liquid cooled optics module that it says will help address the power and performance needed for AI data center network development. With the rapid development of AI, HPC (High-Performance Computing), and 5G, the power density of data centers has increased dramatically. " In this framework, network iteration commences with high-performance GPU scenarios to address premier business demands and subsequently broadens its support to encompass more general scenarios.
Read More
