Keynote: Cloud Native Superpowers with eBPF
summary
Claim: extended Berkeley Packet Filter (eBPF) runs safe, verified programs in the kernel to instrument and change behavior without modifying applications. Demo loads C code via Python bcc attached to execve that prints traces, and a second demo uses XDP drop to block ICMP pings instantly then re-enable them. Verifier prevents unsafe memory or infinite loops. Production uses include kernel-level observability, sidecar-less service mesh and live vulnerability mitigation; performance and correctness measurements show large latency and path savings.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 1:00What is eBPF
- 7:02Why is eBPF so popular
- 10:00Packet of Death
- 15:31The verifier
- 19:09Deep observability
- 21:30Efficient networking
- 24:43Efficiency
- 28:20Service Mesh
- 31:32Cloud Native Tools
related talks
15mKeynote: eBPF: Unlocking Innovation in the Linux Kernel
32mExploring the Power of eBPF for System Observability
40mPanel Discussion: eBPF: A New Era in Cloud Infrastructure Tools
35mAn Introduction to eBPF - Superpowers for Networking, Observability & Security- 40mPanel Discussion: eBPF: A New Era in Cloud Infrastructure Tools
25mWelcome + Opening: Why Everyone is Excited about eBPF in Cloud Native in 2021
43meBPF Superpowers for SRE
37mJust Buzz or Real Byte? eBPF in Cloud Native