Running multiple GPUs per PCIe slot with a switch: inference traffic is tiny
net_termina · x · 2026-10-25
Inference traffic is measured in KB, nowhere near saturating a Gen5 x16 link. The bulk of traffic comes from GPU-to-GPU communication hitting the CPU root complex uplink, which can be avoided entirely with devices like the C-Payne PCIe switch.
Without bifurcation support, each slot can host many GPUs in various inference configurations—as long as you aren't offloading to RAM and P2P is enabled.
More from Infra
- 305M-Parameter Sparse MoE Runs Computer-Use Agent Fully Offline on a MacBook Air — lmoroney · 2026-10-25
- SGLang reruns DeepInfer on matching hardware, matches or beats its prefill performance — hsu_byron · 2026-10-25
- Anthropic in talks for $45B compute deal with Nscale for 460MW of Vera Rubin capacity — Beth_Kindig · 2026-10-25
- DIY GPU cluster war stories: PCIe retimers can literally catch fire — giffmana · 2026-10-25
- GPT-6.1 Sol beats Opus 5.5 on cost-performance balance, argues dev who caught Opus in logic traps — haider1 · 2026-10-25
- Abliterated DeepSeek V4.1 Flash GGUF hits Hugging Face trending — orcarouter · 2026-10-25