Deploy a web application across multiple Availability Zones.
If one AZ becomes unavailable, the ALB must continue sending requests to healthy targets.
Choose TWO appropriate design elements.
Associate the ALB with subnets in multiple AZs and distribute the Auto Scaling targets across those AZs so the entry point and instances survive a single-AZ failure.
Detailed explanation
Sending traffic to unhealthy targets increases the impact of an outage.
Sending traffic to unhealthy targets increases the impact of an outage.
Spreading instances across AZs provides healthy targets outside a failed zone.
Spreading instances across AZs provides healthy targets outside a failed zone.
A single-AZ placement makes the entire web tier vulnerable to that AZ's outage.
A single-AZ placement makes the entire web tier vulnerable to that AZ's outage.
A multi-AZ ALB can continue providing an entry point when one Availability Zone fails.
A multi-AZ ALB can continue providing an entry point when one Availability Zone fails.
Check official documentation
Compare the following points with the official documentation.
Application Load BalancerとEC2 Auto Scalingの複数AZ構成を確認する。
What to confirm
入口とターゲットの両方を複数AZへ配置できる。
Key points
- Application Load Balancer
- Multiple Availability Zones
- Health check
Notes
- Environment: AWS公式ELB・Auto Scalingドキュメントの確認
- Confirm current AWS service behavior in the official documentation.
Foundation review
Two-sided resilience
A multi-AZ ALB alone is insufficient if every target remains in one AZ; both tiers must be distributed.
Failure exclusion
Health checks stop routing to failed targets while the remaining AZs continue serving requests.