WWR2 : IBM Storage Ceph Ready Nodes : Ceph’s Approach : IBM Storage Ceph LIBRADOS API data access
IBM Storage Ceph LIBRADOS API data access
Stores the data with data protection and consistency as the primary goal
Follows a scale-out model to cope with growth and changes
Presents no single point of failure
Provides CRUSH (Controlled Replication Under Scalable Hashing) as a customizable metadata-less placement algorithm
Runs as a software-defined storage solution on commodity hardware
Is open-sourced to avoid vendor lock-in
Figure 271 IBM Storage Ceph Ready Nodes LIBRADOS API data access
Librados and CRUSH are designed to distribute the data across all devices in the cluster to avoid favoring empty devices to write new data.