Regions, zones and who's responsible for what
🎯You'll be able to
Choose a region and spread a service across zones, and explain which security tasks belong to the cloud provider and which belong to you.
🏠The everyday picture
A bank has branches in many cities (regions). In a big city it has several branches in different neighbourhoods (zones), so a power cut in one area doesn't close banking in the whole city. The bank guards the vault; you still have to keep your own PIN secret.
📘What it really is
-
A region is a geographic area with its own data centres, such as AWS Mumbai (
ap-south-1) and Hyderabad (ap-south-2), Google Cloud Mumbai (asia-south1) and Delhi (asia-south2), and Azure Central India (Pune), South India (Chennai) and West India (Mumbai). -
An availability zone is one or more separate data centres within a region, with independent power and networking, connected to the other zones by fast links. Running copies in two or more zones protects against a data centre failure.
-
Choosing a region:
- closeness to users for low latency (Jaipur users are well served from Mumbai or Delhi);
- law and regulation: for example, the Reserve Bank of India requires payment system data to be stored in India, and India's Digital Personal Data Protection Act, 2023, governs how personal data is handled;
-
services and price: not every service is in every region, and prices differ.
-
Multi-region (copies in two regions) protects against a whole-region outage but costs much more and is harder, especially for data. Most services start with multi-zone in one region, plus backups copied to a second region.
-
The shared responsibility model:
- the provider secures the cloud itself: buildings, hardware, the virtualisation layer, and the managed parts of their services;
- you secure what you put in it: who has access, network rules, encryption settings, your operating systems (on virtual machines), your code and your data. The more managed the service, the more the provider takes on, but access and data are always yours.
📝A worked example
Input: map the RouteRani kit onto a cloud design in one region with two zones.
Expected output (a design, not a deployment):
| Kit piece | Cloud equivalent | Spread |
|---|---|---|
| nginx load balancer | managed load balancer (AWS ALB) | across both zones |
API copies (app_count) |
containers on a managed service (ECS on Fargate) | at least one per zone |
| PostgreSQL container + volume | managed database (Amazon RDS for PostgreSQL), Multi-AZ | primary in one zone, standby in the other |
| Docker network | private subnets in a VPC | one per zone |
random_password |
generated password in a secrets manager | regional |
🌍Where it's used
- Exam result and booking sites run in two or more zones so a single data centre problem doesn't take them down on the big day.
- Banks and payment companies pick Indian regions for data residency.
- Disaster recovery plans copy backups to a second region.
📝Case scenario
A ticketing company ran everything in one zone because "the region is reliable". A cooling failure took that zone's servers offline for several hours during a concert sale. Their competitor, with copies in three zones, saw a short slowdown. Afterwards the company ran at least two copies in different zones, with a database standby in another zone.
🖱️Try it
Open the region list for one provider's documentation and find two Indian regions and how many zones each has. Then fill in the table above for Google Cloud (load balancer, Cloud Run, Cloud SQL, VPC, Secret Manager). Success: every kit piece has a named service and a spread.
⚠️Common mistakes
- Everything in one zone. Sign: one data centre problem is a full outage. Fix: copies and database standby in at least two zones.
- Assuming the provider secures everything. Sign: a public storage bucket "because the cloud is secure". Fix: know which settings are yours.
- Choosing a far region for price. Sign: slow pages for Indian users, or data stored where rules don't allow. Fix: weigh latency and law first.
🛠️For your project
In your capstone write-up, include a cloud design table like the one above for RouteRani, with zones and the reasons for your region.
✅Quick check
What's the difference between a region and a zone?
Show the answer
A region is a geographic area; zones are separate data centres within it.
Under the shared responsibility model, who controls access to your data?
Show the answer
You do, whatever service you use.