AWS Fundamental (Load Balancing/ASG/EBS)

Question 1

Vertical Scalability

Answer 1

Increase the size of one instance

Question 2

Horizontal Scalability

Answer 2

Increase the number of instances - ASG/LB

Question 3

High Availability

Answer 3

Running your applications/instances across AZ/regions - ASG/LB multi-AZ

Question 4

Load Balancing

Answer 4

Forward internet traffic to multiple servers

• spreads load across multiple instances
• simple point of entry (DNS)
• seamlessly handle downstream failures

Question 5

ELB

Answer 5

EC2 Load Balancer is a managed load balancer

Question 6

Types of ELB

Answer 6

Classic, Application Load Balancer (ALB), Network Load Balancer (NLB)

Question 7

Application Load Balancers

Answer 7

• Fixed hostname (XXX.region.elb.amazonaws.com)
• Great for microservice & container based solutions
• Can target EC2 instances, ECS tasks, lambda functions and IP Addresses

Question 8

Network Load Balancers

Answer 8

• Forward TCP and UDP traffic
• Can handle millions of requests per second
• Less latency than an ALB
• One static IP per AZ
• Used for extreme performance

Question 9

Load Balancer Stickiness

Answer 9

Redirect clients to the same instance

• ensures users do not lose session data
• can bring imbalance to the EC2 instances

Question 10

Cross Zone Load Balancing

Answer 10

Manage loads across AZ

Question 11

SNI

Answer 11

Server Name Identification - solves the problem of loading multiple SSL certificates onto one web server

Fetches the appropriate SSL certificate based on the hostname requested

Question 12

ELB Connection Draining

Answer 12

When an instance stops, the time allowed to complete a request before deregistering the instance

Question 13

Auto scaling Group

Answer 13

Scale EC2 instances on demand

Question 14

Auto scaling Alarms

Answer 14

Scale ASG based on cloudwatch alarms where an alarm watches a metric (CPU)

• Target Tracking: base it on a simple metric i.e. CPU
• Step/Simple Scaling: More custom, base it a cloudwatch alarm