NVIDIA and ETH Zürich cut small-message AllReduce latency by deleting barriers

thoefler · x · 2026-07-22

NVIDIA and ETH Zürich cut small-message AllReduce latency by deleting a global barrier

The post points to The Barrier Tax, an arXiv paper from NVIDIA and ETH Zürich about long-context decode. The key observation is that once KV cache growth shrinks batch sizes, collectives stop being bandwidth-bound and become synchronization-bound.

What the paper measures

What they changed

Reported gains

Original post →

More from Infra

Infra channel →