{"type":"rich","version":"1.0","provider_name":"Transistor","provider_url":"https://transistor.fm","author_name":"This Week in NET","title":"Inside Cloudflare's Gen 13 Servers: Trading Cache for Cores","html":"<iframe width=\"100%\" height=\"180\" frameborder=\"no\" scrolling=\"no\" seamless src=\"https://share.transistor.fm/e/4797b455\"></iframe>","width":"100%","height":180,"duration":2670,"description":"In this episode of This Week in NET, JQ Lau and Victor Hwang from our Network & Infrastructure Strategy team walk us through Cloudflare's 13th generation of servers — the machines that power a significant part of the internet across 330+ cities worldwide.The Gen 13 program doubled compute density by jumping from 96 to 192 cores, but that came with an 83% drop in L3 cache. The team explains how a bold hardware bet, combined with Cloudflare's FL2 Rust-based software rewrite, turned that trade-off into a win across throughput, latency, and power efficiency.From counterintuitive fan physics to credit card pen tests on chassis intrusion switches, this conversation covers the full stack: CPUs, memory, storage, networking, security, and what's next — including post-quantum readiness at the hardware layer.Mentioned blog posts: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 yetTimestamps00:53 — Blog recap: what Cloudflare announced (including agents can now actually create Cloudflare accounts, buy domains, and deploy)03:52 — From Gen 11 to Gen 13: the evolution of Cloudflare's servers05:04 — Doubling compute power while cutting cache by 83%06:54 — The journey to choosing the right CPU10:04 — Scratchpad vs bookshelf: cache and memory explained12:08 — Why 192 cores won over 128 cores15:35 — FL2: Cloudflare's Rust-based software rewrite18:12 — Hardware and software co-design: why neither works alone18:37 — Memory, storage, and networking upgrades22:18 — Dual GPU support and future accelerators23:25 — Inside the Gen 13 chassis: what changed visually24:51 — Why adding a 5th fan saves power (counterintuitive physics)25:59 — Server security: memory encryption, PCIe encryption, intrusion detection30:12 — 50% better performance per watt and what that means at scale33:54 — The Austin lab: where hardware gets tested before production35:10 — How AI helped design Gen 1337:13 — 500...","thumbnail_url":"https://img.transistorcdn.com/JEk1CAhFcWIBJOFKIzKfdvQ-zUX6aXTuOCJVJJHruOk/rs:fill:0:0:1/w:400/h:400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS9jYThl/MjA0YjY3NjRjMzE4/NTdlZTdjMzFjNjll/YTM1YS5wbmc.webp","thumbnail_width":300,"thumbnail_height":300}