Tackling Container Networking at Scale
summary
Cilium is an eBPF-powered Cilium deployment that runs an operator plus per-node daemonset agents, replaces the CNI (and can replace kube-proxy), and feeds Hubble for packet-level visibility with forwarded/dropped verdicts, port scans and L3/L4/L7 CiliumNetworkPolicies. Installation uses Helm and the cilium CLI; it integrates with Prometheus/alerting and supports multi-cluster and multi-tenancy. At scale it claims deployment on hundreds to thousands of nodes in production (one org reports 10x efficiency) and provides concrete observability and enforcement rather than vague examples.
AI-written summary · about
chapters (11)
- 0:00Container networking intro
- 1:47Scale challenges
- 4:59Current solutions
- 5:41What is Cilium
- 7:52Cilium architecture
- 9:17Security features
- 10:43Cilium scalability
- 11:28Monitoring with Cilium
- 13:15Observability with Hubble
- 14:41Cilium best practices
- 16:06Q&A
related talks
36mCilium: Welcome, Vision and Updates
05mLightning Talk: Unveiling 5 Essential Insights from Cilium Implementation in a...
22mOne Project, Different Angles: How to Secure and Observe with Cilium
16mCilium Wars: Scaling Up Your Network with Jedi-like Cilium Handling Techniques - Yash Tandon
24mThe Cilium Story - Why We Created Cilium
01h 13mCilium Workshop
28mA Decade of Cilium Around the World
31m10 Years of Cilium: Connecting, Securing, and Simplifying the Cloud Nat...