Replication is the process of copying recovery points from one or more source cores and transmitting them to a secondary location/s for the purpose of redundancy and disaster recovery. The process requires a paired source-target relationship between two or more cores.
The source core copies the recovery points of the protected agents and then asynchronously and continuously transmits them to a target core at a remote disaster recovery site. The off-site location can be a company-owned data center (self-managed core) or a third-party managed service provider’s (MSP’s) location or cloud environment. When replicating to a MSP, you can use built-in work flows that let you request connections and receive automatic feedback notifications.
Possible scenarios for replication include:
Replication to a Local Location. The target core is located in a local data center or on-site location, and replication is maintained at regular intervals. In this configuration, the loss of the source Core would not prevent a recovery of protected agents that had been replicated.
Replication to an Off-site Location. The target core is located at an off-site facility or remote data center for recovery in the event of a loss.
Mutual Replication. Two data centers in two different locations each contain a core and are protecting agents and serving as the off-site disaster recovery backup for each other. In this scenario, each core replicates the agents to the Core that is located in the other data center.
Hosted and Cloud Replication. MSP partners maintain multiple target cores in a data center or a public cloud. On each of these cores, the MSP partner lets one or more of their customers replicate recovery points from a source core on the customer’s site to the MSP’s target core for a fee.
Possible replication configurations include:
Point to Point. Replicates a single agent or multiple agents from a single source core to a single target core.
Multi-Point to Point. Replicates multiple source cores to a single target core.
NOTE: The software Core service will replicate all data for a single agent. It is not possible to selectively replicate certain recovery points. All recovery points for volumes that are protected will be replicated when replication is configured.
Use of a seed drive for the initial transfer between source and target cores is recommended. The initial transfer of deduplicated base images and incremental snapshots of the protected agents, can add up to hundreds or thousands of gigabytes of data. Initial replication can be seeded to the target core using external media to transfer the initial data to the target core. This is typically useful for large sets of data or sites with slow links.
NOTE: Dell recommends that you do not use a network connection to seed the base data. Initial seeding involves potentially very large amounts of data, which could overwhelm a typical WAN connection. For example, if the seed data measures 10 GB and the WAN link transfers 24 Mbps, the transfer could take more than 40 days to complete. Because large amounts of data need to be copied to the portable storage device, Dell recommends that you use an eSATA, USB 3.0, or other high-speed connection to the portable storage device.
The data in the seeding archive is compressed, encrypted, and deduplicated. If the total size of the archive is larger than the space available on the removable media, the archive can span across multiple devices based on the available space on the media. During the seeding process, the incremental recovery points are replicated to the target site. After the target core consumes the seeding archive, the newly replicated incremental recovery points automatically synchronize.
Seeding is a two-part process (also known as copy-consume):
The first part involves copying, which is the writing of the initial replicated data to a removable media source. Copying duplicates all of the existing recovery points from the source core to a local removable storage device such as a USB drive. After copying is complete, you must then transport the drive from the source core location to the remote target core location.
The second part is consuming, which occurs when a target core receives the transported drive and copies the replicated data to the repository. The target core then consumes the recovery points and uses them to form replicated agents.
NOTE: While replication of incremental snapshots can occur between the source and target cores before seeding is complete, the replicated snapshots transmitted from the source to the target will remain “orphaned” until the initial data is consumed, and they are combined with the replicated base images.