Overview
Explore the development of scalable NFS clusters using Rook and CephFS in this 37-minute conference talk by Patrick Donnelly from Red Hat. Dive into the integration of Ceph, a distributed storage system, with Kubernetes through Rook. Learn about the dynamic creation of highly-available NFS server clusters that export CephFS for use within Kubernetes or as a standalone appliance. Discover how Ceph and Rook collaborate to manage the deployment of NFS-Ganesha pods with minimal operator involvement. Gain insights into metadata servers, the benefits of Kubernetes and Rook, NFS Ganesha, scale-out architecture, passive-active configurations, NFS clients, grace periods, and cluster file systems. Understand the deployment management process, the relationship between Rook and CephFS, and explore future visions for this technology. The talk concludes with a Q&A session, providing an opportunity to deepen your understanding of this innovative approach to storage management in Kubernetes environments.
Syllabus
Introduction
Metadata Servers
Origin Story
Why Kubernetes
Why Rook
NFS Ganesha
Why Scale Out
Passive Active
NFS Clients
NFS Grace Periods
NFS Over Cluster File System
CephFS Client Reacquire State
NFS Cluster Deployment Management
Rook and CephFS
Summary
Future vision
Questions
Taught by
Linux Foundation