Examkin

Question

Difficulty: MediumManaging Compute Engine Resources

An administrator manages a fleet of non-critical, fault-tolerant web scraping tasks running on Compute Engine instances. The team needs to reduce compute infrastructure costs while collecting detailed host system memory metrics in Google Cloud Observability. Which TWO actions should the administrator take?

  1. Provision the workload instances as Spot VMs to lower compute costs for fault-tolerant tasks.Answer
  2. Install and configure the Google Cloud Ops Agent on the Compute Engine instances.Answer
  3. C
    Install the legacy Cloud Monitoring agent on the VM instances to capture system memory metrics.
  4. D
    Grant the primitive Owner role to the compute instance default service account.
  5. E
    Configure standard persistent disks with Coldline storage class parameters to reduce disk costs.

Answer

The administrator should provision the workload on Spot VMs to optimize cost for fault-tolerant tasks and install the Google Cloud Ops Agent to gather guest system memory metrics.
Provisioning fault-tolerant web scraping tasks as Spot VMs significantly reduces compute costs because these instances take advantage of unused Google Cloud capacity. Additionally, installing the Google Cloud Ops Agent enables the collection of detailed guest OS memory metrics, which are not natively visible at the hypervisor level.

Step-by-Step Solution

1
Evaluate workload fault tolerance for cost optimization
Identify that Spot VMs offer high cost savings for stateless, interruption-tolerant processing.
Spot VMs utilize spare compute capacity at steep discounts.
2
Identify telemetry requirements for memory monitoring
Determine that OS-level memory metrics require an in-guest telemetry agent.
Hypervisor-level metrics cannot inspect guest OS memory without an installed agent.
3
Select the recommended observability agent
Choose the Google Cloud Ops Agent over deprecated legacy agents.
The Ops Agent combines logging and system metric collection under current Google Cloud best practices.

Key Concept

Managing Compute Engine resource efficiency with Spot VMs and gathering guest OS memory telemetry using the Cloud Ops Agent.
Rate this question

Topics

Managing Compute Engine Resources

More questions from this topic