AWS - Elastic Load Balancer - multichoice questions

Question 1

You are designing the architecture for a web application that will be deployed on AWS. The application consists of multiple services running on Amazon EC2 instances. The services communicate over both HTTP and TCP protocols.

Which type of Elastic Load Balancer (ELB) would be most suitable for this scenario?

Application Load Balancer (ALB)
Network Load Balancer (NLB)
Gateway Load Balancer (GWLB)
Classic Load Balancer (CLB)

Question 2

In the security group of an ALB, port 80 is configured to be open for incoming traffic. However, in the security group of the EC2 instances hosting the application, port 80 is not allowed. The instance configured correctly as a target in the ALB.

Given this configuration, would a request to the ALB url, port 80, from a client to the application succeed?

Yes.
No.
Depends on the configuration of the network access control list (NACL) associated with the LB.
Depends on whether the ALB and the EC2 instances are launched within the same availability zone.

Question 3

After experiencing a DDoS attack during the night, you discovered that your instances are accessed directly using their public IP addresses, allowing bypass the ALB that routes traffic to the instance.

How can you overcome this vulnerability (choose all possible)?

Place instances in a private VPC.
Configure the instances to accept traffic only from the ALB using security group rule.
Block any incoming requests to the ALB during nights.
There is no mitigation to this vulnerability.

Question 4

You have set up an ALB to distribute incoming traffic across 2 instances. But when you test how is the load being balanced by performing requests from 15 different regions on the globe, you've noticed that requests are always routed to the same EC2 instance.

Which of the below best match the root cause?

The ALB uses sticky sessions.
The ALB was created in a single AZ.
One of the instances has failed to pass the health checks.
None of the above.

Question 5

You are managing a web service that has two types of requests: short-lived requests that require quick response time, and heavy, long-lived requests that involve significant processing and take a longer time to be completed. The service runs on multiple instances behind an ALB.

There is a risk that the heavy, long-lived requests could potentially monopolize the resources of some instances, causing degradation in the performance and responsiveness of the short-lived requests.

Considering this scenario, which load balancing algorithm would be the preferred choice for distributing traffic to the instances?

Round-robin
Weighted target groups
Lease privilege principle
Least outstanding requests