NVIDIA launches its first co-packaged optics products

The company says its silicon photonics networking switches can scale AI factories to millions of GPUs while dramatically boosting power efficiency and signal integrity
NVIDIA has unveiled NVIDIA Spectrum-X and NVIDIA Quantum-X silicon photonics networking switches, which the company says enable AI factories to connect millions of GPUs across sites while drastically reducing energy consumption and operational costs. The company says it has achieved the fusion of electronic circuits and optical communications at massive scale.
As AI factories grow to unprecedented sizes, networks must evolve to keep pace. According to NVIDIA, its photonics switches are the world’s most advanced networking solution, integrating optics innovations with 4x fewer lasers to deliver 3.5x more power efficiency, 63x greater signal integrity, 10x better network resiliency at scale, and 1.3x faster deployment compared with traditional methods.
“AI factories are a new class of datacentres with extreme scale, and networking infrastructure must be reinvented to keep pace,” said Jensen Huang, founder and CEO of NVIDIA. “By integrating silicon photonics directly into switches, NVIDIA is shattering the old limitations of hyperscale and enterprise networks and opening the gate to million-GPU AI factories.”
NVIDIA silicon photonics networking switches are available as part of the NVIDIA Spectrum-X Photonics Ethernet and NVIDIA Quantum-X Photonics InfiniBand platforms.
According to NVIDIA, the Spectrum-X Ethernet networking platform delivers superior performance and 1.6x bandwidth density compared with traditional Ethernet for multi-tenant, hyperscale AI factories, including the largest supercomputer in the world.
NVIDIA Spectrum-X Photonics switches include multiple configurations, including 128 ports of 800G or 512 ports of 200G, delivering 100T total bandwidth, as well as 512 ports of 800G or 2,048 ports of 200G, for a total throughput of 400T, the company adds.
NVIDIA Quantum-X Photonics switches provide 144 ports of 800G InfiniBand based on 200G SerDes and use a liquid-cooled design to efficiently cool the onboard silicon photonics. The company says the Quantum-X Photonics switches offer 2x faster speeds and 5x higher scalability for AI compute fabrics compared with the previous generation.
NVIDIA’s silicon photonics ecosystem includes TSMC, Browave, Coherent, Corning Incorporated, Fabrinet, Foxconn, Lumentum, SENKO, SPIL, Sumitomo Electric Industries, and TFC Communication.
“A new wave of AI factories requires efficiency and minimal maintenance to achieve the scale required for next-generation workloads,” said C. C. Wei, chairman and CEO of TSMC. “TSMC’s silicon photonics solution combines our strengths in both cutting-edge chip manufacturing and TSMC-SoIC 3D chip stacking to help NVIDIA unlock an AI factory’s ability to scale to a million GPUs and beyond, pushing the boundaries of AI.”
NVIDIA says its photonics will drive massive growth for a new wave of state-of-the-art AI factories, alongside pluggable optical transceiver technologies supported by industry leaders including Coherent, Eoptolink, Fabrinet, and Innolight.
Image credit: NVIDIA