The number of web servers should change with traffic.
Any server should return the same result and a single-instance failure should have limited impact.
Choose TWO appropriate design elements.
An Application Load Balancer distributes requests to multiple EC2 instances, and an Auto Scaling group adjusts the instance count as demand changes.
Detailed explanation
An ALB distributes HTTP(S) requests across targets and provides a redundant entry point.
An ALB distributes HTTP(S) requests across targets and provides a redundant entry point.
A single instance is a single point of failure and cannot horizontally scale.
A single instance is a single point of failure and cannot horizontally scale.
Auto Scaling increases or decreases instances in response to demand.
Auto Scaling increases or decreases instances in response to demand.
Local-only state is not shared and can be lost when a request reaches another instance.
Local-only state is not shared and can be lost when a request reaches another instance.
Check official documentation
Compare the following points with the official documentation.
Elastic Load BalancingとEC2 Auto Scalingの公式説明を確認する。
What to confirm
負荷分散と自動水平スケールを組み合わせたWeb層を説明できる。
Key points
- Application Load Balancer
- EC2 Auto Scaling
- Horizontal scaling
- Stateless web tier
Notes
- Environment: AWS公式ELB・Auto Scalingドキュメントの確認
- Confirm current AWS architecture guidance and service behavior in the official documentation.
Foundation review
Elastic entry
The load balancer spreads requests while Auto Scaling changes the target capacity.
State placement
Shared session or data state should live in a durable shared service rather than one instance disk.