Enabling PCIe P2P on Consumer Nvidia GPUs Boosts LLM Throughput by 25%
BidonPomoev · reddit · 2026-08-09
A developer running VLLM on a server with 4x RTX 5060Ti discovered that enabling PCIe peer-to-peer (P2P) communication on consumer Nvidia cards yields significant performance gains.
Test Setup
- Hardware: AMD EPYC server (150GB/s RAM bandwidth), 4x 5060Ti 16GB in PCIe 4.0 x8
- Model: Qwen/Qwen3.6-27B-FP8 (F16 KV cache, no KV quantization)
- Parallelism: Tensor parallelism
Performance Gains
With P2P enabled, Time To First Token (TTFT) dropped significantly at 32k context depth, and prefill throughput saw a 25% boost for free.
How to Enable
- Hardware must support ReBAR (enable in BIOS).
- Install patched Nvidia community drivers.
More from Infra
- Zero's next-gen IVM engine runs in a single DO, live-querying a 100GiB R2 database — threepointone · 2026-10-16
- Meme: driving to the chip factory in 2028 captures AI's compute anxiety — beffjezos · 2026-10-16
- One volunteer maintains the time zone file behind 4 billion phones — danbri · 2026-10-16
- DeepSeek V4.1 Flash on 4x DGX Spark: weights alone are 511GB, the 494 tok/s claim doesn't fit — Aroochacha · 2026-10-16
- Open-sourced Intel GPU Vulkan inference upgrades: 2x ingestion, 60%+ faster inference — Fun-Aardvark-1143 · 2026-10-16
- ELSFP pain points: >3dB insertion loss, costly PMF, harsh packaging requirements — jwt0625 · 2026-10-16