Intermediate Google Cloud test (Intermediate)

Sample questions

5 sample questions from the Intermediate Google Cloud test, with the answers. Samples are taken out of the live test, so you will not meet these when you sit it.

1. In a GKE project, you need to manage a specific application configuration that is not covered by standard Kubernetes resources. Which of the following approaches would be most appropriate for applying a custom resource definition (CRD) in this scenario?

   1. Use a Kubernetes Operator to manage the application and its configuration, without defining any new resource types.
   2. Use a custom resource definition (CRD) to define a new resource type for the application-specific configuration and manage it through Kubernetes APIs.
   3. Use a ConfigMap to store the application-specific configuration and reference it in the Kubernetes deployment.
   4. Use a Helm chart to package the application and its configuration, and deploy it to GKE.

   Answer: B. Use a custom resource definition (CRD) to define a new resource type for the application-specific configuration and manage it through Kubernetes APIs.

   In a realistic project scenario, a custom resource definition (CRD) is used to define a new resource type that is specific to the application. This allows developers to manage application-specific configurations through Kubernetes APIs, making it easier to integrate with the Kubernetes ecosystem and automate operations.

2. What is the correct method to upgrade a specific node pool in GKE to a desired version?

   1. Upgrade the entire GKE cluster using `gcloud container clusters upgrade` without specifying a node pool.
   2. Use the `gcloud container clusters upgrade` command with the `--node-pool` flag to upgrade the specific node pool to the desired version.
   3. Use `kubectl` to upgrade a specific node pool by specifying the node pool name in the command.
   4. Use the GKE web console to manually upgrade a specific node pool to the desired version.

   Answer: B. Use the `gcloud container clusters upgrade` command with the `--node-pool` flag to upgrade the specific node pool to the desired version.

   To upgrade a specific node pool in GKE, the correct method is to use the `gcloud container clusters upgrade` command with the `--node-pool` flag, which allows you to target the specific node pool for the version upgrade. Other options such as using `kubectl` or upgrading the entire cluster without specifying the node pool are not accurate or may not achieve the desired result. Upgrading the entire cluster without targeting a specific node pool is not the correct approach when only a specific node pool needs to be upgraded.

3. Which statement accurately describes the configuration differences between dead-letter topics and push subscriptions in Pub/Sub?

   1. Dead-letter topics are configured with an HTTP endpoint for message delivery, while push subscriptions are configured with a maximum number of delivery attempts.
   2. Both dead-letter topics and push subscriptions are configured with a maximum number of delivery attempts.
   3. Dead-letter topics are configured with a maximum number of delivery attempts, while push subscriptions are configured with an HTTP endpoint for message delivery.
   4. Both dead-letter topics and push subscriptions are configured with an HTTP endpoint for message delivery.

   Answer: C. Dead-letter topics are configured with a maximum number of delivery attempts, while push subscriptions are configured with an HTTP endpoint for message delivery.

   Dead-letter topics in Pub/Sub are used to handle messages that have failed to be processed after a specified number of delivery attempts. They are configured with a maximum number of delivery attempts, which determines how many times a message is retried before being moved to the dead-letter topic. On the other hand, push subscriptions are configured with an HTTP endpoint, which is the destination where messages are pushed. These are two distinct configurations, and the correct statement accurately describes the differences between them.

4. Which of the following statements accurately describes how the GKE cluster autoscaler operates?

   1. The cluster autoscaler scales based on the number of deployed applications.
   2. The cluster autoscaler scales individual pods based on network traffic.
   3. The cluster autoscaler scales node pools based on metrics like CPU utilization and memory usage.
   4. The cluster autoscaler scales based on the number of active users.

   Answer: C. The cluster autoscaler scales node pools based on metrics like CPU utilization and memory usage.

   The GKE cluster autoscaler automatically adjusts the size of node pools based on metrics such as CPU utilization and memory usage to meet the demands of the workloads running in the cluster. This ensures optimal resource utilization and performance. Other options are incorrect because the autoscaler does not scale based on network traffic, does not scale individual pods, and does not scale based on the number of deployed applications.

5. Which method of data ingestion into BigQuery is best suited for real-time processing and ensures minimal latency in data delivery?

   1. Batch loading using CSV files
   2. Using Cloud Data Transfer Service
   3. Streaming ingestion using Apache Kafka
   4. Using Cloud Storage as a staging area

   Answer: C. Streaming ingestion using Apache Kafka

   BigQuery can ingest data in real-time using streaming ingestion, which is often implemented through Apache Kafka or other messaging systems. This method allows data to be written to BigQuery as it is generated, ensuring low-latency ingestion and enabling real-time analytics. Other options like batch loading, using CSV files, or data transfer services are valid methods but do not represent the real-time or streaming aspect of data ingestion.