Cloud deployment models

Teach

A cloud deployment model describes where your resources live and who owns the infrastructure.

The three deployment models

• Public cloud runs on shared infrastructure owned by the provider (Azure). No hardware to buy, fast to scale, pure pay-as-you-go. Resources are logically isolated per customer.
• Private cloud uses infrastructure dedicated to a single organization — on-premises or hosted. Maximum control and isolation, but you own maintenance and capacity.
• Hybrid cloud combines public and private, letting workloads and data move between them. Good for gradual migration, regulatory data residency, and bursting to the cloud for peak demand.

A quick comparison: public = lowest cost and effort, private = most control, hybrid = flexibility to keep sensitive workloads on-premises while using the cloud for the rest.

Hybrid connectivity

To make hybrid work, you connect on-premises networks to Azure:

• VPN Gateway creates an encrypted site-to-site VPN over the public internet — quick to set up, lower cost, variable bandwidth.
• ExpressRoute provides a private, dedicated connection through a connectivity provider — higher, more consistent bandwidth, lower latency, and traffic that never touches the public internet.

Quick decision checkpoints

1. Want minimal cost and operations? Choose public.
2. Need strict control or data must stay on-premises? Choose private.
3. Need both, or migrating gradually? Choose hybrid with VPN or ExpressRoute.

Pick the model from your control, compliance, and cost needs; then choose the right connectivity for hybrid.