White Papers

12 PS Series Asynchronous Replication Best Practices and Sizing Guide | BP1012
3 PS Series replication process
When a replica is created, the first replication process completes the transfer of all volume data. For
subsequent replicas, only the data that changed between the start time of the previous replication cycle and
the start time of the new replication cycle is transferred to the secondary group. Dedicated volume snapshots
are created and deleted in the background as necessary to facilitate the replication process. Logically, you
could describe a volume replica set as a combination of the following:
Volume
replica
set
=
A full copy of the primary volume, with
data synchronized to the beginning of the
most current completed replication.
+
A time-sequenced set of replicas, in which
each replica corresponds to the state of the
volume at the beginning of a prior replication.
The number of prior replicas that are stored on the secondary group is limited by the size of the replica
reserve allocated for that volume and the amount of data that changes. See section 7.4 to properly size a PS
Series replication solution. The replication process can be described as a series of phases. The flowchart in
Figure 4 shows the process phases, focusing on how the process tracks and copies changes that occur
between each replica cycle.
The primary group checks for availability of sufficient delegated and replica reserve space on the secondary
group at the beginning of each replication processing phase. If adequate space is not available, the process
will pause and generate an event message. Replication will continue once sufficient space is made available.
These parts of the process are not shown in the chart.
Proper sizing and capacity planning should be considered before the replication solution is decided. Section
7,
Best practices for planning and design, provides direction on the proper capacity as well as performance
considerations for asynchronous replication with PS Series storage.
The following subsections refer to the phases shown in Figure 4.
3.1 Replication setup (one-time)
This phase configures the replication partnership and volume replication settings.
3.2 Replication processing (repeating)
Primary-to-secondary volume data replication is completed in this phase. The process steps vary based on
replication status (first or subsequent) and fast failback mode (enabled or disabled). During this process, the
local reserve is consumed by a hidden snapshot (and the fast failback snapshot if enabled). Volume data
changes that occur during the replication processing phase are stored by the hidden snapshot in the local
reserve. The replica reserve allocated to the volume within delegated space on the secondary group receives
all volume data changes. The replica reserve is consumed by the most recent complete volume replica plus
all prior replicas stored in the replica set.