What's changed: Revamped CLF-C02 Chapter 2 to the new depth (Regions/AZs/edge/containment/selection, SPOF/multi-AZ+ELB+Auto Scaling/HA/fault tolerance/elasticity/DR)
2.1Regions, Availability Zones, and Edge Locations
Understand AWS’s geography—Regions, Availability Zones (AZs), and edge locations—and how to choose a Region for your resources.
AWS is built from datacenters worldwide, and where you place resources is a more important design decision than it appears. It directly affects latency, legal/data-residency (data sovereignty), cost (varies by Region), and the availability of the services you want. This section organizes AWS geography—Region > Availability Zone > datacenter—plus the separate edge locations.
2.1.1Regions and Availability Zones
- Region: a geographically separate area (e.g., Tokyo ap-northeast-1), made up of multiple AZs (usually three or more).
- Availability Zone (AZ): one or more isolated datacenters with independent power, cooling, and networking, interconnected over low-latency dedicated links.
- Spreading across AZs (multi-AZ) makes you resilient to a single AZ failure.
The key is that AZs are physically separated, placed far enough apart that a local incident (lightning, flood, power loss) is unlikely to hit multiple AZs at once. That is why "spread across AZs and survive one failure" holds (design covered next). For wide-area disasters, consider backup/replication to another Region.
2.1.2Edge locations and related infrastructure
- Edge locations: sites that cache/deliver content near users (Amazon CloudFront); more numerous than AZs, used for low-latency delivery.
- Local Zones: place compute/storage near large metros for ultra-low-latency needs (supplementary).
- The four Region-selection factors: latency (proximity), compliance/data residency, cost (varies by Region), and service availability.
| Element | Scope | Main role |
|---|---|---|
| Region | Geographic area | Where you place resources |
| Availability Zone | Isolated DCs within a Region | Multi-AZ for high availability |
| Edge location | Delivery sites worldwide | Low-latency delivery via CloudFront |
| Factor | Why it matters |
|---|---|
| Latency (proximity) | Closer to users = faster response |
| Compliance/data residency | Where data may legally reside |
| Cost | Same service priced differently by Region |
| Service availability | New services launch in some Regions first |
Scenario: domestic video delivery. Viewers are mostly in-country → choose the Tokyo Region (latency). Cache static videos at edge locations (CloudFront) for low-latency global delivery. For availability, run the origin across multiple AZs, and back up to another Region for wide-area disasters.
Watch the mix-ups: (1) AZ (isolated DC within a Region, for HA) vs edge location (CloudFront delivery site, for low latency)—different roles. (2) "multi-AZ" is redundancy within one Region; wide-area disasters need another Region. (3) Region choice is not about preference—decide by latency, compliance, cost, and availability.
Common points: AZ = isolated DC within a Region (multi-AZ for redundancy) / edge location = CloudFront content delivery / Region selection = latency, compliance, cost, service availability. Don’t confuse AZ with edge locations.
2.1.3Section summary
- Containment: Region > AZ > datacenter; AZs are physically separated
- Edge locations = low-latency delivery near users via CloudFront (a different role than AZs)
- Region selection = latency, compliance/data residency, cost, service availability
Sign in to track progress — Log in.
Quick check
(just a quick review)Q1. Which correctly describes an Availability Zone (AZ)?
Q2. Which caches/delivers content near users for low latency?
Q3. Which is NOT a relevant consideration when choosing a Region?
Q4. Which containment is correct?
Q5. What is the basic placement to withstand a single datacenter failure within a Region?