AI Dynamics

Global AI News Aggregator

About

Hedged Requests Reduce P99.99 DRAM Read Latency

Hedged requests (apparently inspired by the Tail at Scale paper by myself and Luiz Barroso) applied within a single machine to replicating data across DRAM channels and issuing reads to all channels, using the one that comes back first. ~5-15X reduction in p99.99 read latency.

→ View original post on X — @jeffdean