@no_stp_on_snek: Cisco coming back into relevance again! Let's gooo
Summary
NVIDIA 和 Cisco 联合推出可插拔光学模块（USFP），首次支持每通道 400Gb/s，带宽密度提升 4 倍，帮助 AI 数据中心在不增加网络层级的情况下连接更多算力并降低单位功耗。
View Cached Full Text
Cached at: 10/13/26, 05:36 AM
Cisco coming back into relevance again! Let’s gooo
NVIDIA Newsroom (@nvidianewsroom): NVIDIA and @Cisco introduce pluggable optics to scale AI factories.
USFP is the first module to support 400Gb/s per lane, delivering 4x the bandwidth density of today’s optics. AI builders can connect more compute to the same switch — no extra network tiers, less power per bit
Similar Articles
@CNET: Nvidia CEO Jensen Huang talks about NVIDIA Co-Packaged Optics at GTC 2026
Supermicro and NVIDIA unveiled turnkey “AI Factory” racks at GTC 2026, offering pre-validated, rack-scale compute/storage/networking stacks to speed enterprise AI deployment.
Csco state of ai by Dotadda
Cisco is repositioning itself as a core AI infrastructure supplier, reporting strong hyperscaler orders and a broadening market beyond GPUs to networking, security, and observability. The company raised its FY26 hyperscaler AI infrastructure order expectation to ~$9B and highlighted Silicon One as a critical differentiator.
Coherent Breaks Ground on Expanded Texas Facility, Scaling AI’s Optical Backbone
Coherent breaks ground on an expanded facility in Texas, backed by a $50 million CHIPS Act grant, to scale production of optical components and compound semiconductors essential for AI infrastructure. NVIDIA CEO Jensen Huang highlighted the critical role of optics in enabling large-scale AI systems.
China’s optical chip breakthrough boosts AI speed 100-fold using fraction of compute power
Chinese researchers have developed an all-optical interconnect system that links standard electronic chips, boosting AI distributed inference speeds by over 100 times while using just one-ninth of typical computational resources. The breakthrough, published in National Science Review, uses silicon photonic transceiver chips and FPGAs to achieve dramatic efficiency gains.
NVIDIA Spectrum-X — the Open, AI-Native Ethernet Fabric — Sets the Standard for Gigascale AI, Now With MRC
NVIDIA's Spectrum-X Ethernet fabric with Multipath Reliable Connection (MRC) sets a new standard for gigascale AI networking, improving throughput, load balancing, and resilience for large-scale AI training, as demonstrated by OpenAI, Microsoft, and Oracle.