Andromeda: Performance, Isolation, and Velocity at Scale in Cloud Network Virtualization
summary
Andromeda is Google Cloud’s network virtualization stack that uses a hierarchy of controllers plus Hoverboard gateways to avoid programming the long tail of mostly idle flows and offloads high‑bandwidth VM pairs once they exceed ~100 kb/s (results: >99.5% of traffic bypasses hoverboards). On hosts an OS bypass fast path plus per‑VM coprocessors handles packets; a single core hits ~38 Gbps and round‑trip latency ranges ~25 μs (15 μs w/ busy polling, ~9 μs with DPDK). Design scales to 100,000 VMs per network; 40k‑VM programming time falls from ~67s to ~2s; upgrades cause ~270 ms median blackout.
AI-written summary · about
chapters (10)
- 0:00Intro
- 1:20Network Virtualization
- 2:18Andromeda Architecture
- 6:02Scaling with Hoverboards
- 6:40Hoverboard Offloading
- 9:24Andromeda Data Plane
- 11:28Data Plane - Fast Path
- 13:09Data Plane - Coprocessor Path
- 16:33CPU Efficiency Evolution
- 18:56Dataplane Hitless Upgrade 1/3
related talks
11mBuilding Resilient Network Infrastructure
34mHyperRouter: Lessons Learnt from Building an L4 Load...
36mPanel Disucssion: Is There Actually a Byte Behind All the Buzz? eBPF in Production!
21mZero Overhead Monitoring for Cloud-native Infrastructure using RDMA
34mThe Missing Metrics: Measuring Memory Interference in Cloud Native Systems
34mTroubleshooting Low Latency Application on CNF Deployment
31mCloud Migration Without Fear - A...
13mGraviton2: How Honeycomb Reduced Infra Spend by 40%
