Concurrency and scaling in GCP - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ service_behavior

intermediate

How does Google Cloud Run handle concurrency by default?

Google Cloud Run allows you to configure concurrency for your container instances. What is the default concurrency setting for Cloud Run services?

A Each container instance can handle up to 80 concurrent requests by default.

B Each container instance handles exactly 1 request at a time (concurrency = 1).

C Each container instance can handle unlimited concurrent requests by default.

D Each container instance handles 10 concurrent requests by default.

💡 Hint

Think about how Cloud Run optimizes resource use by handling multiple requests per container.

By default, Cloud Run sets concurrency to 80, meaning each container instance can handle up to 80 requests at the same time. This helps optimize resource usage and reduce cold starts.

❓ Architecture

intermediate

What happens when a Google App Engine Standard environment instance reaches its maximum concurrent requests?

In Google App Engine Standard environment, each instance can handle a limited number of concurrent requests. What does App Engine do when an instance reaches this limit?

A App Engine queues the extra requests until the instance can handle them.

B App Engine automatically creates new instances to handle the extra requests.

C App Engine rejects the extra requests with an error response.

D App Engine throttles the requests by slowing down the instance.

💡 Hint

Think about how App Engine scales to handle more traffic.

When an instance reaches its concurrency limit, App Engine automatically creates new instances to handle additional requests, enabling automatic scaling.

❓ security

advanced

Which practice improves security when scaling Google Kubernetes Engine (GKE) clusters under heavy load?

When your GKE cluster scales up to handle more pods, which practice helps maintain security effectively?

A Allow all pods to run with root privileges to avoid permission issues during scaling.

B Use a single node pool with wide permissions for all workloads to simplify management.

C Disable network policies to improve pod communication speed during scaling.

D Use Pod Security Policies or Pod Security Admission to restrict pod permissions consistently.

💡 Hint

Think about how to keep pods secure even as new ones are added automatically.

Using Pod Security Policies or Pod Security Admission ensures that all pods, including new ones created during scaling, follow strict security rules, reducing risk.

❓ Configuration

advanced

How to configure autoscaling in Google Compute Engine Managed Instance Groups (MIG) based on CPU usage?

You want your MIG to add or remove VM instances automatically based on CPU load. Which configuration snippet correctly sets autoscaling to target 60% average CPU utilization?

A gcloud compute instance-groups managed set-autoscaling my-group --max-num-replicas=10 --min-num-replicas=1 --target-cpu-utilization=0.6

B gcloud compute instance-groups managed set-autoscaling my-group --max-num-replicas=10 --min-num-replicas=1 --target-cpu-utilization=60

C gcloud compute instance-groups managed set-autoscaling my-group --max-num-replicas=10 --min-num-replicas=1 --cpu-utilization-target=60

D gcloud compute instance-groups managed set-autoscaling my-group --max-num-replicas=10 --min-num-replicas=1 --cpu-threshold=0.6

💡 Hint

Check the correct flag name and value format for CPU utilization target.

The correct flag is --target-cpu-utilization and the value is a decimal between 0 and 1 representing the target CPU usage fraction.

🧠 Conceptual

expert

Why is horizontal scaling often preferred over vertical scaling in cloud environments?

Consider a web application running in the cloud. Why do architects usually prefer horizontal scaling (adding more machines) instead of vertical scaling (adding more power to one machine)?

A Vertical scaling allows unlimited resource increase without downtime.

B Vertical scaling is cheaper and faster to implement but less reliable.

C Horizontal scaling provides better fault tolerance and can handle more traffic by distributing load.

D Horizontal scaling requires specialized hardware, making it less flexible.

💡 Hint

Think about what happens if one machine fails in each scaling approach.

Horizontal scaling adds more machines, so if one fails, others keep working. It also spreads traffic, improving performance and reliability. Vertical scaling has limits and can cause downtime.