Google Professional Cloud Network Engineer Question 437
Select 2Google Cloud PlatformYour company hosts a web application on Google Cloud using a managed instance group (MIG) behind a global external HTTP(S) load balancer. The application experiences unpredictable traffic spikes, and you want to ensure that it can handle sudden workload increases without manual intervention while remaining cost-effective. What steps should you take to configure the environment?
- A
Configure an autoscaling policy with a target CPU utilization threshold for the managed instance group.
- B
Set the minimum number of instances to match the expected peak traffic volume.
- C
Enable autoscaling for the managed instance group and set a maximum number of instances.
- D
Disable autoscaling and manually adjust the number of instances during traffic spikes.
- E
Configure the load balancer to send traffic only to a fixed number of instances.
Show answer and explanation
Correct answers: A, C
Explanation
To handle unpredictable traffic spikes efficiently, you should enable autoscaling on the managed instance group. Configuring an autoscaling policy with a target CPU utilization ensures that the number of instances automatically adjusts based on workload demands. Setting a maximum number of instances helps maintain cost control while ensuring the application can accommodate traffic surges. Manual scaling and fixed configurations are not practical or cost-effective solutions for handling dynamic workloads.
- A. Correct.
Correct: Configuring an autoscaling policy with a target CPU utilization allows the managed instance group to automatically adjust the number of instances based on actual workload, ensuring efficient resource usage.
- B. Incorrect.
Incorrect: Setting the minimum number of instances to match peak traffic volume is not cost-effective, as it keeps unneeded resources running during low-traffic periods.
- C. Correct.
Correct: Enabling autoscaling and setting a maximum number of instances ensures the application can scale to handle traffic spikes while also preventing excessive scaling that could lead to unnecessary costs.
- D. Incorrect.
Incorrect: Disabling autoscaling and manually adjusting the number of instances is not recommended, as it requires continuous monitoring and intervention, which is error-prone and inefficient.
- E. Incorrect.
Incorrect: Configuring the load balancer to send traffic to a fixed number of instances limits the scalability of the application and does not accommodate workload increases effectively.