NVDA 145.98 ▼3.30%GOOGL 272.66 ▼3.46%MSFT 498.30 ▼1.78%AMD 489.19 ▼7.72%INTC 55.72 ▲1.92%TSMC 352.09 ▼2.90%AMZN 257.37 ▼3.87%META 596.22 ▼3.27%AAPL 349.21 ▼0.92%PLTR 153.68 ▼1.91%
Markets at last close

Infrastructure

Lambda deploys NVIDIA optical switches for large GPU clusters

·1 min read

Lambda is using NVIDIA Photonics co-packaged optics switches to connect NVIDIA GB300 NVL72 racks with more than 10,000 GPUs. The deployment targets lower networking power consumption, greater resilience and longer intervals between interruptions, which can stall training jobs or cause cascading delays in agentic inference.

In Lambda’s tests, network latency dropped by about 4% at rack scale and by up to about 10% at pod scale. Throughput remained the same as conventional pluggable optics across rack, scalable unit and pod configurations. NVIDIA says its technology delivers up to 5x better power efficiency and up to 15x higher reliability than conventional pluggable optics, while using 4x fewer lasers.

Each switch has 144 ports running at 800Gb/s, with liquid cooling removing heat from the switch chip and optical engines. Lambda reports a median optics temperature of 36°C, compared with 51°C for conventional pluggable optics, alongside more stable signal integrity in testing.

The design places optical engines beside the switch chip, shortening electrical paths and eliminating retimer chips to reduce signal loss, power draw and component count. Each switch also has 18 removable external light sources that technicians can replace in the field, rather than replacing an entire optical module when a laser fails.

Originally reported by blogs.nvidia.comRead the source →
Related coverage