Implementation of the backup strategy with Managed Veeam Backup
It has been determined that a total of 6 VMs are required to maintain regularly scheduled backups. For reasons of cost and efficiency, the university's IT department has chosen to make these backups using the "Veeam Managed Backup" solution.
The alternative considered more expensive and complicated to set up, was to opt for a storage server, connected to the infrastructure through the vRack. This would require a minimum investment of $109/month and more time to being spent configuring the machine with putting in place the Rsync protocol, verifying and testing backups, monitoring and maintaining the server and having to worry about inevitable hardware failures.
“Veeam Managed Backup” is a simple solution: it’s an option, which can be activated within the OVH customer control panel, then the user only has to select the VM Backup by right-clicking within the vSphere hypervisor. Backups are made on a space which is independent of the of the user’s Dedicated Cloud (mounted RAID repository, hosted and managed by OVH). In other words, there is a backup solution as a Service. Data passes through a management network (VLAN; the backup does not affect the performance of the service's network. Another advantage, the space required to backup data is not deducted from the user’s personal Dedicated Cloud space. The only visible aspect for the user is a Windows VM, which is light on resources (1 vCPU), running on their Dedicated Cloud. The VM is created by OVH.
The billing of the service is transparent: $14/month/backed up VM, no matter the size. In our case, it would be $84/month (in comparison the cost of $109/month for a dedicated backup server at OVH.com, not to mention the time spent by the university’s IT team to configure and maintain a dedicated backup server).
A daily backup, with a snapshot of the VM which the automatic backup was requested is sent to the repository. The backup retention period is between 14 and 20 days, this for so the user can restore data to the same state as several days prior to an incident. Used for example when a virus infects a service several days ahead of administrators becoming aware of any issues.
In practice, Veeam Managed Backup performs a complete weekly backup of the VM, plus 12-18 incremental backups. As a consequence, "full" backups may take longer to perform (this is dependent on the amount of data the VM contains) whereas backups from previous days, only contain the changes or addition to files since the last backup. Each day, a daily report is sent to the infrastructure’s administrator to inform him of the successful completion of backups.
In the event of an incident on the Dedicated Cloud infrastructure, therefore the university’s social network can restart with a maximal data loss of 24 hours. This RPO (Recovery Point Objective) can be attained with the use of a configuration management tool.This allows for, after the restoration of the vital VMs, the replication of the multi-tier architecture of production.
Refer to the complete users guide for Veeam Managed Backup