Why load balancing matters in GCP - Test Your Understanding

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create a load balancer that distributes traffic evenly.

GCP

gcloud compute forwarding-rules create my-load-balancer --load-balancing-scheme=[1] --ports=80 --target-pool=my-pool

Drag options to blanks, or click blank then click option'

AEXTERNAL

BDYNAMIC

CSTATIC

DINTERNAL

Attempts:

3 left

💡 Hint

Common Mistakes

Choosing INTERNAL when traffic is from the internet.

The load balancing scheme should be EXTERNAL to distribute traffic from the internet to backend instances.

2fill in blank

medium

Complete the code to specify the protocol used by the load balancer.

GCP

gcloud compute forwarding-rules create my-load-balancer --load-balancing-scheme=EXTERNAL --ports=[1] --target-pool=my-pool

Drag options to blanks, or click blank then click option'

A443

B80

C22

D3306

Attempts:

3 left

💡 Hint

Common Mistakes

Using port 443 which is for HTTPS.

Port 80 is used for HTTP traffic, which is common for web load balancing.

3fill in blank

hard

Fix the error in the backend service creation command to enable health checks.

GCP

gcloud compute backend-services create my-backend --protocol=HTTP --health-checks=[1] --global

Drag options to blanks, or click blank then click option'

Amy-health-check

Bdefault-check

Chealth-check-1

Dcheck-health

Attempts:

3 left

💡 Hint

Common Mistakes

Using a health check name that does not exist.

The health check must match the name of the created health check resource, here my-health-check.

4fill in blank

hard

Fill both blanks to create a firewall rule allowing load balancer health checks.

GCP

gcloud compute firewall-rules create allow-health-checks --direction=INGRESS --priority=1000 --network=default --action=ALLOW --rules=[1] --source-ranges=[2]

Drag options to blanks, or click blank then click option'

Atcp:80

Bicmp

C130.211.0.0/22

D0.0.0.0/0

Attempts:

3 left

💡 Hint

Common Mistakes

Allowing all IPs (0.0.0.0/0) which is insecure.

Health checks use TCP port 80 and come from the IP range 130.211.0.0/22.

5fill in blank

hard

Fill the blanks to configure a backend service with session affinity and timeout.

GCP

gcloud compute backend-services create my-backend --protocol=HTTP --session-affinity=[1] --timeout-sec=[2] --health-checks=my-health-check --global

Drag options to blanks, or click blank then click option'

ACLIENT_IP

B30

C60

DNONE

Attempts:

3 left

💡 Hint

Common Mistakes

Using too high timeout causing delays.

Session affinity is set to CLIENT_IP to keep user sessions consistent, timeout is 30 seconds.

Practice

1. Why is load balancing important in cloud infrastructure?

easy

A. It stores data permanently for backup purposes.

B. It encrypts data to secure communication.

C. It distributes incoming traffic evenly across multiple servers to prevent overload.

D. It monitors user activity for analytics.

Solution

1. Step 1: Understand the role of load balancing

   Load balancing helps distribute user requests evenly to multiple servers to avoid any single server from becoming overwhelmed.

2. Step 2: Compare with other options

   Options A, C, and D describe different cloud functions like storage, encryption, and monitoring, which are not related to load balancing.

3. Final Answer:

   It distributes incoming traffic evenly across multiple servers to prevent overload. -> Option C

4. Quick Check:

   Load balancing = traffic distribution [OK]

Hint: Load balancing spreads traffic to avoid server overload [OK]

Common Mistakes:

• Confusing load balancing with data storage
• Thinking load balancing encrypts data
• Assuming load balancing is for analytics

2. Which of the following is the correct way to configure a basic HTTP load balancer in GCP?

easy

A. Create a backend service, add instance groups, and set up a URL map.

B. Create a Cloud Storage bucket and upload files.

C. Enable Cloud Functions and write a trigger.

D. Set up a BigQuery dataset and run SQL queries.

Solution

1. Step 1: Identify load balancer components

   In GCP, an HTTP load balancer requires a backend service, instance groups as backends, and a URL map to route traffic.

2. Step 2: Eliminate unrelated options

   Options A, B, and C relate to data analytics, storage, and serverless functions, not load balancing setup.

3. Final Answer:

   Create a backend service, add instance groups, and set up a URL map. -> Option A

4. Quick Check:

   Load balancer setup = backend + instance groups + URL map [OK]

Hint: Load balancer needs backend service and URL map [OK]

Common Mistakes:

• Confusing load balancer setup with storage or functions
• Skipping URL map configuration
• Not adding instance groups as backends

3. Consider this simplified code snippet configuring a load balancer backend in GCP:

backend_service = create_backend_service()
add_instance_group(backend_service, 'group-1')
add_instance_group(backend_service, 'group-2')

requests = 100
handle_requests(backend_service, requests)

What is the expected behavior when 100 requests arrive?

medium

A. Requests are sent to a single instance in 'group-2' only.

B. All 100 requests go to 'group-1' only.

C. Requests are dropped because no URL map is set.

D. Requests are evenly distributed between 'group-1' and 'group-2'.

Solution

1. Step 1: Understand backend groups in load balancing

   When multiple instance groups are added to a backend service, the load balancer distributes incoming requests evenly among them.

2. Step 2: Analyze request handling

   With 100 requests and two groups, requests will be balanced roughly 50 each, not all to one group or dropped.

3. Final Answer:

   Requests are evenly distributed between 'group-1' and 'group-2'. -> Option D

4. Quick Check:

   Multiple backends = balanced requests [OK]

Hint: Multiple backends share traffic evenly [OK]

Common Mistakes:

• Assuming all traffic goes to one group
• Thinking requests drop without URL map
• Believing only one instance gets all requests

4. You set up a load balancer but notice all traffic is going to a single server, causing overload. What is the most likely cause?

medium

A. The health checks are passing for all instances.

B. The backend service has only one instance group attached.

C. The load balancer is configured with multiple backend groups.

D. The URL map is routing traffic evenly.

Solution

1. Step 1: Identify cause of uneven traffic

   If all traffic goes to one server, likely only one instance group is attached, so no distribution occurs.

2. Step 2: Check other options

   Multiple backend groups or even URL map routing would distribute traffic. Healthy instances mean traffic can be balanced.

3. Final Answer:

   The backend service has only one instance group attached. -> Option B

4. Quick Check:

   Single backend = no traffic distribution [OK]

Hint: One backend group means no load balancing [OK]

Common Mistakes:

• Assuming multiple backends cause overload
• Ignoring health check status
• Believing URL map causes single-server traffic

5. A company wants to ensure their web app stays available even if one server fails. They have three servers behind a load balancer. Which setup best achieves this goal?

hard

A. Use a load balancer with health checks and three backend servers in different zones.

B. Use a single server with a backup database.

C. Use three servers without a load balancer and rely on DNS round robin.

D. Use a load balancer but disable health checks to avoid false failures.

Solution

1. Step 1: Understand high availability requirements

   To stay available if one server fails, traffic must shift automatically to healthy servers in different zones.

2. Step 2: Evaluate options for failover

   Use a load balancer with health checks and three backend servers in different zones. uses load balancer with health checks and multi-zone servers, ensuring automatic failover and availability.

3. Step 3: Reject other options

   Use a single server with a backup database. lacks redundancy; Use three servers without a load balancer and rely on DNS round robin. lacks health checks and proper failover; Use a load balancer but disable health checks to avoid false failures. risks undetected failures by disabling health checks.

4. Final Answer:

   Use a load balancer with health checks and three backend servers in different zones. -> Option A

5. Quick Check:

   Health checks + multi-zone = high availability [OK]

Hint: Health checks + multi-zone servers ensure uptime [OK]

Common Mistakes:

• Disabling health checks thinking it helps
• Relying on DNS round robin alone
• Using single server for high availability