Google Professional Cloud DevOps Engineer Question 266
Select 3Google Cloud PlatformYour company runs multiple applications on Google Cloud, each requiring varying levels of compute resources throughout the month. Most workloads are predictable and consistent, while others experience spikes in usage during specific periods. To optimize infrastructure costs, which strategies should you consider implementing?
- A
Purchase committed-use discounts for predictable workloads with consistent resource usage.
- B
Rely exclusively on preemptible VMs for all workloads to reduce costs significantly.
- C
Leverage sustained-use discounts for workloads with varying but long-running usage patterns.
- D
Enable network tiers to reduce egress costs for traffic with specific latency requirements.
- E
Use autoscaling to dynamically allocate resources for spiky workloads.
Show answer and explanation
Correct answers: A, C, E
Explanation
To optimize infrastructure costs, you should combine strategies that align with your workload patterns. Committed-use discounts are effective for predictable, consistent workloads, while sustained-use discounts automatically reduce costs for long-running, variable workloads. Autoscaling is crucial for handling spiky workloads efficiently. While preemptible VMs and network tiers offer cost-saving opportunities, they are not universally applicable to all workloads or directly tied to the scenario in question.
- A. Correct.
Committed-use discounts are ideal for predictable workloads with consistent usage patterns, as they provide significant cost savings in exchange for a long-term commitment.
- B. Incorrect.
While preemptible VMs are cost-effective, they are not suitable for all workloads due to their ephemeral nature and lack of guaranteed availability, making them unsuitable for critical or consistent workloads.
- C. Correct.
Sustained-use discounts automatically apply for long-running workloads with varying usage patterns, helping reduce costs without needing upfront commitments.
- D. Incorrect.
Network tiers optimize egress costs based on latency requirements, but they are not directly related to compute cost optimization or infrastructure cost planning as described in this scenario.
- E. Correct.
Autoscaling helps dynamically allocate resources for workloads with spiky or unpredictable usage, ensuring efficient resource utilization while minimizing unnecessary costs.