Basic Information
| Field | Details |
|---|---|
| English name | Elastic Load Balancing |
| Full name | Elastic Load Balancing |
| Chinese description | 弹性负载均衡 |
| Japanese description | Elastic Load Balancing（マネージド負荷分散サービス） |
| Exam frequency | ⭐⭐⭐⭐⭐ |
| Often confused with | Auto Scaling / ALB / NLB / GWLB |
In one sentence
Distributes traffic across healthy targets to improve availability and scale application entry points.
Key points
- ALB operates at Layer 7 for HTTP/HTTPS routing by host, path, and application rules.
- NLB operates at Layer 4 for very high-performance TCP/UDP/TLS traffic and static IP needs.
- GWLB inserts and scales virtual network appliances.
Exam takeaway
Load balancing distributes traffic; Auto Scaling changes target capacity.
Update: listeners, TLS, and target state
- A listener defines the accepted protocol and port; a rule combines priority, conditions, and actions; a target group defines backends and health checks.
- SNI selects a certificate during the TLS handshake, host rules select target groups at HTTP time, and ALPN negotiates an application protocol.
- Registered, Healthy, and InService are distinct states; an Active load balancer does not prove application availability.
- Deregistration delay lets in-flight work finish before removal, while stickiness never overrides health checks.