Ceph is a software-defined distributed storage system with high fault tolerance. Its main advantages are:
— accessing data in object, block and file storage systems; — open source nature that reduces operating costs, ensures fast development of Ceph, provides support for the professional community; — advanced backup and data integrity algorithms, recovery tools that maintain services and systems in case of emergencies.
Who This Course Is For?
Engineers already working with Ceph
You will get a better understanding of Ceph and fill your knowledge gaps. You can also ask specific questions about your infrastructure.
The new technology will expand your expertise. Ceph is now gaining popularity sparking the demand for competent specialists.
Specialists who prioritize fault tolerance
Ceph architecture facilitates data migration and updates, helps to build a fault-tolerant infrastructure.
You will have access to all modules in your Slurm Account.
№1: What Ceph Is and Is Not
What Ceph Is
What Ceph can and cannot do
Alternatives to Ceph
№2: Ceph Architecture Overview
Ceph components' functionality
№3: How to Install Ceph
Ceph-deploy, cephadm, ansible
System requirements in a nutshell
№4: Ways to Use Ceph
External interfaces: RBD, CephFS, RGW (S3)
№5: Ceph Integration with Common Cloud Native Solutions
№6: Exploitation of Ceph
Replace/add/remove OSD, MDS, RGW
Migrate a monitor to another server
Manage configurations. Where to find and how to change them
Server reboot. How to stop a cluster
№7: Monitoring Ceph
Collect metrics. What to pay attention to
№8: How to Fix Problems with Ceph
Global flags/flapping OSD. How to stop rebalancing
Data overflow in clusters
№9: Ceph Performance. Simple Math
What Is Ceph cluster/storage performance
What your cluster can and cannot do
How to estimate and calculate performance
What you need to do to reach a certain performance level
№10: How to Choose Hardware for Your Cluster
General recommendations on choosing hardware for your cluster
№11: Ceph Pools and Storage Classes
Meet Our Course Speakers
— Expert developer at CUSTIS, Linuxoid, Ceph specialist — Codes in React, Node.js, PHP, Go, Python, Perl, Java, C ++, designs infrastructure solutions — Considerable expertise in Ceph performance — DevOps engineer with 7 years of experience and a co-founder of DevOps Engineers of Saint-Petersburg community
— DevOps consultant, former cloud infrastructure specialist — Open-source storage systems enthusiast — Specializes In infrastructure support, development in Java, Go, TypeScript, JS, PHP, Perl, C, C ++ — Once saw a Ceph storage cluster running two monitors
— Lead engineer in cloud development group at CROC — Has been successfully using Ceph for over 6 years, 24/7 — On top of Ceph, has experience in distributed storage systems like Dell EMC VxFlex (ScaleIO), IBM Spectrum Scale (GPFS), Gluster — Develops storage and virtualization subsystems at CROC