Google Professional Cloud DevOps Engineer Question 171
Select 4Google Cloud PlatformYour organization has recently adopted Google Cloud Operations Suite (formerly Stackdriver) to monitor a fleet of microservices running on Google Kubernetes Engine (GKE). However, the team is concerned about the rising costs of observability. As a Google Professional Cloud DevOps Engineer, what actions can you take to manage and optimize observability costs while maintaining effective monitoring?
- A
Reduce the data retention period for logs in Cloud Logging.
- B
Filter out unnecessary logs using exclusion filters in Cloud Logging.
- C
Disable monitoring for all services to completely eliminate observability costs.
- D
Leverage sampled tracing in Cloud Trace to reduce the amount of trace data collected.
- E
Use aggregated metrics instead of high-cardinality metrics in Cloud Monitoring.
Show answer and explanation
Correct answers: A, B, D, E
Explanation
Observability costs can increase significantly with increased data volume and retention. To manage these costs effectively, you can optimize log retention, filter unnecessary logs, sample trace data, and use aggregated metrics. Completely disabling observability is not a viable solution as it hinders your ability to monitor and troubleshoot your systems effectively.
- A. Correct.
Reducing the data retention period for logs in Cloud Logging can help lower costs as storage expenses decrease with shorter retention durations.
- B. Correct.
Exclusion filters in Cloud Logging allow you to avoid storing logs that are not needed, directly reducing costs associated with log storage.
- C. Incorrect.
Disabling monitoring for all services is not a feasible solution as it compromises observability, which is critical for troubleshooting and ensuring service reliability.
- D. Correct.
Sampled tracing in Cloud Trace reduces the volume of trace data collected and stored, which can help control costs while still providing essential trace insights.
- E. Correct.
Using aggregated metrics reduces the storage and processing overhead of high-cardinality metrics, which can significantly lower monitoring costs.