Managing Cloud Native Artifacts for Large Scale Kubernetes Cluster
Talkat KubeCon + CloudNativeCon North America 2020 Virtual · on Friday, 2020-11-20· at 17:55· Intrado Virtual Event Platform,·
summary
Demonstrates running Harbor as a scale-out registry with Harbor 2.1 proxy cache, cross-data-center replication, non‑blocking garbage collection and RBAC, plus vulnerability scanning that can block pulls. Describes NetEase operating 20+ Harbor instances, largest with ~100,000 images across clusters of 5,000+ nodes, solving distribution bottlenecks by preheating P2P networks using Dragonfly/Kraken and a shared registry mode to reach ~5,000 concurrent peers and 10+ GB/s distribution acceleration. Presents concrete HA, storage and CI/CD integration choices.
AI-written summary · about
chapters (17)
- 0:00Intro
- 0:41Agenda
- 1:08Cloud Native Apps Management
- 2:08Harbor for Artifact Management
- 3:01Artifact replication
- 4:04Proxy cache project
- 5:23High Availability (HA) and Scalable Registry
- 6:20High Availability(HA) and Scalable Registry (2)
- 7:40Non-Blocking Garbage Collection
- 8:30P2P Preheat
- 11:02Security Considerations
- 11:50Enhanced Artifact Processor
- 13:13Scenarios in NetEase
- 14:07Our architecture
- 16:13Used in multi-environments
- 19:19Integrated with CI/CD
- 21:21The Harbor Authoritative Guide
related talks
28mHarbor - Enterprise Cloud Native Artifact Registry
43mDeep Dive: Harbor - Enterprise Cloud-native Artifact Registry
41mIntro: Harbor - Enterprise Cloud Native Artifact Registry
21mEnterprise and Harbor: A Good Partnership in Helping Each Other
33mHarbor - Enterprise Cloud Native Artifact Registry
34mDeep Dive: Harbor
34mIntro: Harbor
36mHarbor