AWS region codes reference
An AWS Region is a physical cluster of data centres in one geographic area, each containing multiple isolated Availability Zones (AZs). Picking the right region affects latency for your users, data-residency compliance obligations, which services are available, and price (costs vary by region). This reference maps every commercial AWS region code — like us-east-1 or ap-southeast-2 — to its location, continent, AZ count, and opt-in status.
How region codes are structured
Every region code encodes three parts:
<geography>-<direction>-<number>- Geography prefix:
us(United States),eu(Europe),ap(Asia Pacific),sa(South America),ca(Canada),af(Africa),me(Middle East),il(Israel). - Sub-area:
east,west,central,north,south,southeast,northeast. - Launch-order number: reflects when the region launched in that sub-area, not any ranking.
So ap-northeast-1 = Asia Pacific, north-east area, first region → Tokyo. eu-west-1 = Europe, west area, first region → Ireland.
Opt-in vs always-on regions
| Status | Examples | Behaviour |
|---|---|---|
| Always enabled | us-east-1, eu-west-1, ap-northeast-1 | Available on all accounts by default |
| Opt-in required | ap-east-1 (HK), me-south-1 (Bahrain), af-south-1 (Cape Town), eu-south-1 (Milan) | Must enable in Account Settings before deploying resources |
Regions launched after approximately March 2019 are opt-in by default. When you enable an opt-in region, IAM and some other global services need additional setup, and not every AWS service is available from day one in new regions.
Why us-east-1 is special
us-east-1 (N. Virginia) is the oldest and largest AWS region. Several global control-plane endpoints — IAM, Route 53, CloudFront, AWS Organizations, AWS Billing — resolve to us-east-1 regardless of which region you are working in. Always keep us-east-1 enabled, even if none of your production workloads run there.
Choosing a region for your workload
Three factors typically drive the choice:
- Latency to users — deploy in the region geographically closest to the majority of your traffic.
- Data residency — GDPR requires personal data on EU residents to stay in the EU (use
eu-*regions). Other regulations have similar requirements. - Service availability — some newer or specialised services launch in a subset of regions first; check the service endpoint list before committing.
For high-availability production deployments, prefer regions with three or more AZs — most major regions have three, and the largest have six. Multi-AZ deployments survive a complete data centre failure without downtime.
GovCloud and China regions
This reference covers the standard commercial partition (aws). AWS GovCloud (aws-us-gov) and the China regions (aws-cn) are separate partitions with isolated accounts and different service endpoints. They are noted separately rather than mixed into the commercial list.
Common region mistakes to avoid
Hardcoding a region in client configuration is the most common issue. If you set region: "us-east-1" in code and deploy a service to users in Europe, every SDK call makes a transatlantic round trip. Use an environment variable or infrastructure-level configuration so the region can be overridden per deployment environment.
Forgetting to enable opt-in regions before deploying causes confusing errors. If a Terraform plan references ap-east-1 but the region has not been enabled in the AWS account, resources fail to create with cryptic permission errors. Enable the region in Account Settings first, wait a few minutes for propagation, then deploy.
Assuming all services are available in all regions. AWS launches services incrementally. A new service may be live in us-east-1 months before it reaches ap-south-1. Always check the service availability table in the AWS documentation before choosing a region based on a specific service requirement.
Using region codes in user-visible strings. Region codes like eu-central-1 are internal identifiers, not user-facing labels. For user interfaces, map the code to a human-readable name such as “Europe (Frankfurt)” — this reference provides those display names alongside each code.
Latency benchmarks by region pair
Round-trip latency between AWS regions varies significantly. As a rough order of magnitude: within the same continent (for example us-east-1 to us-west-2) expect 60–80 ms. Cross-Atlantic (for example us-east-1 to eu-west-1) is typically 80–100 ms. Cross-Pacific can exceed 150 ms. These figures depend on internet routing conditions and fluctuate; for production use, measure actual latency from your deployment target rather than relying on estimates.