3.3. Planning and Implementing Networking Resources
💡 First Principle: Google Cloud's networking model is built around one foundational, frequently-tested fact — the VPC network itself is global, while the subnets inside it are regional — and nearly every other networking decision (connectivity, load balancing, service tiers) is really a consequence of designing around that split.
Get the global/regional split wrong in your mental model, and networking scenarios stop making sense: how can resources in us-east1 and europe-west1 share the same VPC without a VPN between them? Because the VPC isn't region-bound in the first place — only its subnets are.
⚠️ Common Misconception: A VPC network is scoped to a single region, the way it works in most other clouds. In reality, a Google Cloud VPC network is a global resource; its subnets are the regional resources within it — a distinction that changes how you plan multi-region connectivity from the ground up.
