Administrator Guide
Figure 50. Local Replication
Replication can be used in various scenarios to achieve dierent levels of data protection.
Replication Scenarios Description
Fast backup and restore Maintains full copies of data for protection against data loss, corruption, or user mistakes
Remote data access Applications can access mirrored data in read-only mode, or in read-write mode if NAS volumes
are promoted or cloned
Online data migration Minimizes downtime associated with data migration
Disaster recovery Mirrors data to remote locations for failover during a disaster
Conguring replication is a three-step process:
• Add a replication partnership between two FluidFS clusters.
• Add replication for a NAS volume.
• Run replication on demand or schedule replication.
How Replication Works
Replication leverages snapshots. The rst time you replicate a NAS volume, the FluidFS cluster copies the entire contents of the NAS
volume. For subsequent replication operations, the FluidFS cluster copies only the data that changed since the previous replication
operation started. This design allows for faster replication, ecient use of system resources, and saves storage space while keeping data
consistent. Replication is asynchronous, meaning that each source NAS volume can have a unique schedule for replicating data to the
target NAS volume.
The amount of time replication takes depends on the amount of data in the NAS volume and the amount of data that has changed since
the previous replication operation.
When replicating a NAS volume to another FluidFS cluster, the other FluidFS cluster must be set up as a replication partner. Each FluidFS
cluster can have multiple replication partners, enabling you to replicate dierent NAS volumes to dierent partners, depending on
operational requirements. However, each individual NAS volume can be replicated to only one target NAS volume on one replication partner.
The following gure summarizes which replication scenarios are supported.
FluidFS Administration
459