3. Reference design for backup services

This chapter will describe a complete Veeam Cloud Connect Backup deployment at a service provider. By impersonating a provider willing to offer backup services, you will learn about design and deployment of all the servers, networks and network rules necessary to run Veeam Cloud Connect Backup.

3.1: Cloud Connect Backup overview

All the components will be deployed in VMs running on top of a hypervisor in order to leverage the quick deployment times of new VMs starting from templates and to protect components that cannot be executed in multiple instances, like the Veeam Backup & Replication server. Because of this, sizing rules will be based on vCPUs (virtual CPUs) rather than sockets and cores. Additionally, the size adjustments of virtual disks will be easier than in a physical server, like when it will be needed to increase the WAN accelerator cache size for example.

The only exception is backup repositories: For better performance, service providers should use physical backup repositories.

During this chapter (and the following one dedicated to replication services), these VMs are referred to as servers; please remember that physical servers only refer to the repositories.

Note: If you plan to use physical servers, adjust the CPU considerations to existing and available CPU models and specifications.

---

Table of contents

• 3.1 Network diagram for backups
• 3.2 Security zones
• 3.3 Veeam Cloud Connect Backup deployment
• 3.4 Cloud gateways
• 3.5 Cloud gateway pools
• 3.6 WAN accelerators
• 3.7 Cloud repositories
• 3.8 Immutable backups
• 3.9 Tenants to Tape