ELB & ASG Flashcards
INCOMPLETE
1
Q
Why use a load balancer?
A
- Spread load across multiple instances
- Expose a single point of access (DNS)
- Seamlessly handle server failures
- Regular health checks
- SSL termination (HTTPS)
- High availability across zones
2
Q
ALB
A
Application Lode Balancer
- HTTP / HTTPS / gRPC protocols
- Layer 7
- HTTP routing
- Statis DNS (URL)
2
Q
What are the 4 AWS load balancers?
A
- Application load balancer (HTTP/HTTPS) layer 7
- Network load balancer layer 4
- Gateway load balancer layer 3
- Classic load balancer (retired, Layer 4 & 7)
3
Q
NLB
A
Network load balancer
- TCP / UDP
- layer 4
- high performance (millions req/s)
- Static IP through elastic IP
4
Q
GLB
A
Bateway load balancer
- GENEVE protocol on IP packets
- Layer 3
- Route traffic to firewalls on EC2 instances
- Intrusion detection
5
Q
ASG
A
Auto scaling group
- Add instances (scale out)
- Remove instances (scale-in)
- Ensure min and max instances
- Automatically register new instances to load balancer
- Replace unhealthy instances
6
Q
Three autoscaling strategies
A
- Manual scaling
- Dynamic scaling
- Predictive scaling
7
Q
Properties of dynamic scaling
A
- Simple / step scaling
- Add two when CPU > 70%
- Remove one when CPU < 30%
- Target tracking scaling
- ASG CPU average at 40%
- Schedule scaling
- based on known usage patterns
- increase min capacity for peak hours
8
Q
Properties of predictive scaling
A
- Use ML to predict future traffic
- Automatically provision the right number of EC2 instances
- useful when usage patterns are predictable
