Accelerates Image Distribution In Multi-Cluster With Dragonfly
Talkat KubeCon + CloudNativeCon North America 2022 · on Friday, 2022-10-28· at 14:00· Virtual Platform Only·
summary
Dragonfly is a peer-to-peer image and file distribution system; the talk explains its system architecture, technology choices used in its design, and the implementation of its intelligent scheduling. It reviews operational problems encountered at large scale and the specific solutions applied, gives best practices for accelerating container image distribution across multiple Kubernetes clusters, and describes integrations with ecosystem tools like Harbor and Nydus.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 4:13Architecture
- 9:57Feature Upgrades
- 11:34Seed Peer
- 13:30Keepalive
- 15:37Consistent Hashing
- 17:25Filter And Evaluator
- 19:57Directed Acyclic Graph
- 23:28TLS Transportation
- 24:31Dfstore
- 25:23Multi-Cluster Dragonfly
- 26:42Acceleration Framework For Image
related talks
27mDragonfly: Make Image Distribution Efficiently and Safely in Cloud Native
31mThe New Evolution of Image Distribution System Dragonfly
34mDeep Dive: Dragonfly
37mIntro: Dragonfly
36mDragonfly: Intro, Updates and AI Model Distribution in the Practice of...
29mPull Once, Scale Everywhere: Fixing Image Pull Bottlenecks With Dragonfly
22mManaging Cloud Native Artifacts for Large Scale Kubernetes Cluster
37mCache Me If You Can: P2P Image Sharing in Kubernetes with Spegel