Learning objectives

• Deploy Ceph and integrate it with OpenStack
• Use Ceph storage services effectively
• Operate and maintain the environment
• Scale the cluster as demand grows
• Monitor health and performance
• Troubleshoot common issues
• Capacity planning
• Backup and recovery concepts
• Best practices

Ceph

Ceph is open-source, software-defined storage that runs on a cluster of standard servers and disks. It stores data in a distributed way, with automatic replication, so data remains safe if a server or disk fails.

Ceph can present storage as:

• Block storage (like virtual disks)
• File storage (like a shared file system)
• Object storage (like S3 or Swift)

Scale out almost without limits by simply adding more servers and disks.

OpenStack

OpenStack is open‑source cloud software that lets you build and manage your own cloud infrastructure (like AWS or Azure, but on your own servers). It provides:

• Compute (VMs, containers)
• Storage (block, object)
• Networking and security
• Authentication, roles, and project isolation
• Image and snapshot management
• Orchestration and automation

You can use it to run workloads, virtual machines, and services in a private or hybrid cloud.

Why Ceph + OpenStack

• Unified storage for cloud services — Ceph supplies block (RBD), object (RGW), and file (CephFS) storage that OpenStack services (Cinder, Glance, Nova, Swift) can all use, simplifying architecture and operations.

• Scalability — both projects scale horizontally: add compute nodes for OpenStack and add OSDs/nodes for Ceph to grow capacity and throughput without major rework.

Why Ceph + OpenStack

• High availability and resilience — Ceph’s replication/erasure coding and self‑healing ensure data stays available if disks or nodes fail, so OpenStack VMs and volumes remain reliable.

• Operational efficiency — one converged storage platform reduces the number of different storage systems to manage, lowering operational overhead and troubleshooting surface.

Why Ceph + OpenStack

• Performance and flexibility — Ceph can be tuned per workload (replication, erasure coding, cache tiers), letting you match cost/performance for VM disks, images, and object workloads.

• Automation and integration — both expose APIs and integrate well with automation tools (cephadm, Ansible, OpenStack APIs), enabling repeatable deploys, upgrades, and CI/CD workflows.

Why Ceph + OpenStack

• Cost control and openness — open‑source stack running on commodity hardware reduces licensing costs and avoids vendor lock‑in; you can choose hardware and optimizations that fit your budget.

• Feature synergy — Ceph features (RBD snapshots/clones, RGW S3 semantics, CephFS shared storage) enable OpenStack features like fast volume provisioning, image storage, backups, and shared file systems for cloud apps.

• OpenStack and Ceph are both part of the OpenInfra ecosystem — built to work together for best interoperability and community-driven integration.

Course roadmap

• Day 1: Architecture, Preparations and Setup, Ceph Deployment
• Day 2: Ceph Operations, Scale Out, Monitroing, Troubleshootint
• Day 3: OpenStack Deployment
• Day 4: OpenStack Operations, Scale Out, Mounitoring Troubleshooting
• Day 5: Backup and Recovery, Automation, Best Practices

Basic concepts

• Instances (VMs), Flavors, Quotas
• Images, Volumes, Snapshots
• Controller, Network, Compute
• Networks, subnets, routers, security groups, and provider vs tenant networks.
• Projects and Roles

Basic concepts

• Mon and OSD Nodes
• Block Storage(RBD), Object Storage(RGW, S3), CephFS
• CRUSH map, Placement Groups (how Ceph decides where data lives)

OpenStack Core Components

• Keystone
• Nova
• Neutron
• Glance
• Cinder

Ceph Core Components

• MON (monitor)
• OSD (object storage daemon)
• MGR (manager)
• MDS (metadata server)
• RGW (object gateway).

Prerequisites

• Linux adminstration
• Networking foundamentls
• Storage basics
• Virtualization concepts
• Troubleshooting skills
• Configuration management

Workshop

• Hands-on - Everyone builds their own environment
• Autonomous - Everyone gains their own practical experience
• Open - All slides, labs and code are available online
• Reproducible - Applicable to any (standard) hardware
• Adaptable - Content can be adjusted for skill level and environment.

Guidelines

• 45/15 Split: 45m sessions | 15m coffee & catch-up
• Questions - please raise your hand to ask questions
• Punctuality - please return promptly after breaks
• Applied Learning - Practice over Theory