Analysis of the EPYC 145% performance gain in Cloudflare Gen 12 servers
Oct 15, 2024JQ Lau, Syona Sarma
Summary
This post details the sensitivity analysis of Cloudflare's primary workload on AMD EPYC 4th generation processors (Genoa, Bergamo, Genoa-X). It quantifies performance gains from core count (linear scaling), TDP (incremental improvements up to 400W), and frequency. Crucially, it highlights the significant impact of L3 cache size, with the Genoa-X variant (9684X) showing a 2.75x performance multiplier in lab simulations over the Milan 7713 (Gen 11), suggesting Cloudflare's software stack is highly sensitive to L3 cache. The analysis uses the internal cf_benchmark tool and a performance simulation tool to evaluate CPU candidates, providing concrete metrics and trade-offs.
More from the Hardware Infrastructure & Performance Optimization trail
- How we reduced core unit boot time from hours to minutes
- A one-line Kubernetes fix that saved 600 hours a year
- Launching Cloudflare’s Gen 13 servers- trading cache for cores for 2x edge compute performance
- Inside Gen 13- how we built our most powerful server yet
- So long, and thanks for all the fish- how to escape the Linux networking stack