Veeam Backup & Replication provides disaster recovery through image-based VM replication. The SP can expose resources of their virtualization environment to tenants as cloud hosts.

Tenants can utilize cloud hosts provided by the SP to create VM replicas offsite. In case of a disaster on the production site, tenants can quickly and easily switch to VM replicas in the cloud and use the SP infrastructure as a remote disaster recovery site.

The SP can provide Veeam Cloud Connect Replication resources for the following virtualization platforms:

VMware vSphere
Microsoft Hyper-V
As well as a cloud repository, the Veeam Cloud Connect Replication infrastructure has a multi-tenant architecture. The SP allocates computing, storage, and network resources for a replication target and provides them to tenants through hardware plans. For the SP, a hardware plan is an abstraction layer that lets the SP virtually partition a virtualization host or cluster into multiple replication targets. As a result, the SP can expose replication resources to several tenants and store tenants’ data in the cloud in an isolated and segregated way.

For a tenant, a hardware plan appears as a cloud host that can be used as a regular target host for off-site replication.

To make VM replicas on the cloud host accessible over the network after failover, Veeam Backup & Replication provides every tenant with network resources — network extension appliances and dedicated VLANs. The tenant can fail over a group of production VMs (full site failover) or individual VMs (partial site failover) to VM replicas on the cloud host. Veeam Backup & Replication establishes a secure channel between VM replicas in the cloud and VMs on the production site and offers traffic encryption capabilities.

In case of software or hardware malfunction on the production site, a tenant can quickly recover a corrupted VM by failing over to its replica in the cloud. When you perform cloud failover, a replicated VM on the cloud host takes over the role of the original VM. A tenant can fail over to the latest state of a replica or to any of its good known restore points.

Veeam Cloud Connect Replication supports failover and failback operations for one VM and for several VMs. In case one or several hosts fail, you can use batch processing to restore operations with minimum downtime.

Depending on the scale of the disaster that affects the production site, a tenant can choose one of the following cloud failover scenarios:

          * Full site failover — the whole production site becomes unavailable and all critical VMs that run interdependent applications fail over to their replicas on the cloud host.
          * Partial site failover — one or several VMs become corrupted and fail over to their replicas on the cloud host.

In Veeam Backup & Replication, the actual failover is considered a temporary stage that should be further finalized. While the replica is in the Failover state, you can undo failover, perform failback or perform permanent failover.

If a tenant wants to resume operation of a production VM, he or she can fail back to it from a VM replica on the cloud host. When you perform failback, you get back from the VM replica to the original VM, shift your I/O and processes from the cloud host to the source production host and return to the normal operation mode.

A tenant can perform failback to a production VM after partial site failover or full site failover. If a tenant performs the failback operation after full site failover, he or she needs to process every VM in the cloud failover plan individually.

If a tenant managed to restore operation of the source host at the production site, a tenant can switch from the VM replica to the original VM on the source host. If the source host is not available, a tenant can restore the original VM to a new location and switch back to it. To learn more, see the Replica Failback section in the Veeam Backup & Replication User Guide.

Failback to production is a temporary stage that should be further finalized. After a tenant tests the recovered original VM and make sure it is working without problems, he or she should commit failback. A tenant also has an option to undo failback and return the VM replica back to the Failover state.

Tenants who plan to replicate their VMs to the cloud must configure replication jobs on their Veeam backup servers and target them at the cloud host. When a job starts, Veeam Backup & Replication performs the following actions:

          1. The tenant starts a replication job. The Veeam backup server on tenant’s side sends a request to the cloud gateway to access the cloud host.

          2. The cloud gateway passes this request to the SP Veeam backup server.

          3. The SP Veeam backup server provides a TLS certificate and establishes a secure connection between the SP Veeam backup server and tenant Veeam backup server.

          4. VM data from tenant’s side is transported via the cloud gateway to the cloud host. If the SP has several cloud gateways, VM data is transported via the least loaded cloud gateway being online.