This section describes the Veeam Backup for Microsoft Office 365 architecture in more details.
For enhanced scalability and load balance in distributed environments, Veeam offers a new scalable architecture of Veeam Backup for Office 365, as shown in the figure below:
- Veeam Backup for Microsoft Office 365 server – the central management component that controls licenses and global settings (list of exclusions, email notifications) for the backup proxies. The RestfulAPI service also runs on that server.
- Proxies – the components that perform data backup. A backup proxy can run on a physical or a virtual machine. Initially, the default backup proxy is installed on the Veeam Backup for Microsoft Office 365 server during the product setup, and then users can deploy more proxies to optimize data processing. Each proxy can process one or several Exchange organizations; an organization can be processed by one or several proxies (this means “many-to-many” connections between proxies and organizations).
If a Veeam Backup for Microsoft Office 365 server fails, its managed proxies will wait for connection to resume for 48 hours. During this period, they will go on with data backup, so that if the server is up and running again in 2 days, all backups will become available via the management console.
- Repository – a folder, local or shared (experimental mode), where backup data is stored. Each repository is associated with a backup proxy. Note that you can configure more than one repository for a proxy. For example, a proxy can back up HQ mailboxes to the Repository 11, and remote office mailboxes – to the Repository 12.
- Microsoft Exchange Organization – an Exchange Online, on-premises or hybrid organization whose mailbox database should be backed up. An organization can be processed by one or several backup proxies at a time.
- Job – a collection of configuration settings for organization data backup. Thus, a job represents a backup policy for the organization mailboxes: it prescribes what mailboxes to backup, what data to exclude, what repository storage to use (from those associated with the proxy), and specifies data collection schedule.
For example, in the figure above you can see:
- Two jobs processed by a single proxy but using different repositories (Job 11 and Job 12).
- Two jobs processed by a single proxy (Job 21 and Job 22), both storing data to a single repository.
- A simple configuration scenario where an organization is processed by only one job using a single repository (Job 31)
Scenarios with multiple jobs and multiple repositories may be recommended for service providers.
It may be reasonable to configure backup jobs for the organization considering geographical locations of ROBOs, administrative boundaries, business units and so on.
Organizations and their backup jobs are displayed in the Organizations view of the product UI. Backup proxies and their associated repositories are displayed in the Backup Infrastructure view of the UI. See the Getting to Know the User Interface section for details.