---

Table of contents

1. TODO
2. The Well-Architected Review’s reliability section currently takes you through the following questions:
3. The relationship between AWS Region, AWS AZs, VPC and subnets
   1. Visualizing the Relationship
   2. Why this matters (Networking & High Availability)

---

TODO

1. How to create a Hub-and-Spoke Topologies. Use the tutorial to build a real structure in the AWS to understand the concepts of VPC, Subnet, AWS Transit Gate Way and CIDR blocks, EC2, Security groups, Loading balancer, etc.

The Well-Architected Review’s reliability section currently takes you through the following questions:

1. REL1 – how do you manage service quotas and constraints?
2. REL2 – how do you plan your network topology?
3. REL3 – how do you design your workload service architecture?
4. REL4 – how do you design interactions in a distributed system to prevent failures?
5. REL5 – how do you design interactions in a distributed system to mitigate or withstand failures?
6. REL6 – how do you monitor workload resources?
7. REL7 – how do you design your workload to adapt to changes in demand?
8. REL8 – how do you implement change?
9. REL9 – how do you back up data?
10. REL10 – how do you use fault isolation to protect your workload?
11. REL11 – how do you design your workload to withstand component failures?
12. REL12 – how do you test reliability?
13. REL13 – how do you plan for disaster recovery?

The relationship between AWS Region, AWS AZs, VPC and subnets

The following example comes from DeepSeek.

Visualizing the Relationship

AWS Region: us-east-1 (N. Virginia)
│
├── Availability Zone: us-east-1a
│   │
│   └── VPC: vpc-12345 (CIDR: 10.0.0.0/16)
│       │
│       ├── Subnet A (10.0.1.0/24) --> (Lives ONLY in us-east-1a)
│       └── Subnet B (10.0.2.0/24) --> (Lives ONLY in us-east-1a)
│
├── Availability Zone: us-east-1b
│   │
│   └── VPC: vpc-12345 (Same VPC as above)
│       │
│       ├── Subnet C (10.0.3.0/24) --> (Lives ONLY in us-east-1b)
│       └── Subnet D (10.0.4.0/24) --> (Lives ONLY in us-east-1b)
│
└── Availability Zone: us-east-1c
    │
    └── VPC: vpc-12345 (Same VPC as above)
        │
        ├── Subnet E (10.0.5.0/24) --> (Lives ONLY in us-east-1c)
        └── Subnet F (10.0.6.0/24) --> (Lives ONLY in us-east-1c)

Why this matters (Networking & High Availability)

You cannot put a subnet in two AZs. If you want an application to survive an AZ failure (e.g., a power outage in us-east-1a), you must:

Create two subnets in the same VPC.

Place Subnet 1 in us-east-1a.

Place Subnet 2 in us-east-1b.

Launch EC2 Instance A in Subnet 1.

Launch EC2 Instance B in Subnet 2.

Put an ELB (Load Balancer) in front of both.

If AZ a dies, traffic routes to Instance B in AZ b.