Active File Management (AFM) can be used for Disaster Recovery (DR) solutions and there are two Spectrum Scale clusters or sites in this case, the primary and secondary.
The primary site exclusively creates or writes data and hence is the “Active” site. The secondary site cannot modify the data and hence is the “Passive” site.
The AFM modes useful for DR are called primary and secondary filesets. AFM DR has a mandatory 1:1 mapping between each primary and secondary fileset. This ensures that only a particular primary, can talk to a secondary.
In normal circumstances, the applications write data on the primary site and this incremental data gets pushed to the secondary site.
In the figure above, you can see that data is moving from primary to secondary site.
In the event of a disaster when the primary site fails or become temporarily unavailable (E.g. for maintenance or planned downtime) then the applications can be moved to the secondary site for continued availability. This process is called as “Failover”. It is shown by the dotted path in the figure.
Before moving the applications, the secondary needs to be promoted, to act like a primary. This helps to ensure business continuity in the event of a disaster with minimal downtime and minimal data loss.
Once the primary site comes back, “Failback” can be done and applications can be moved back to the primary. After, this the relationship continues back in the old manner.
AFM DR moves data continuously and asynchronously in the background and does not rely on snapshots, to do same.
If a disaster happens and the primary site is lost then a new primary can be established using the existing secondary site.
If the secondary site gets destroyed then a new secondary site can be established for the existing primary.
The swapping of the primary and secondary sites can also be done if required.
Please see the below link for more details: