Examkin

Soru

Zorluk: ZorPlanning Compute Engine Resources and Machine Types

A healthcare research organization runs daily genomics analysis batch jobs on Google Cloud. Each batch job execution requires an uninterrupted 4-hour run window and cannot tolerate instance preemptions. Resource profiling indicates that each worker task requires precisely 10 vCPUs and 20 GB of memory to run efficiently while minimizing per-vCPU software licensing costs. Which Compute Engine resource planning strategy meets these operational requirements at the lowest cost?

  1. Deploy standard Compute Engine VM instances using a custom machine type configured with 10 vCPUs and 20 GB of memory.Cevap
  2. B
    Deploy Spot VM instances using a custom machine type configured with 10 vCPUs and 20 GB of memory.
  3. C
    Deploy the genomics processing workloads using Cloud Functions with maximum memory and execution timeout allocations.
  4. D
    Deploy predefined n2-standard-16 VM instances covered by 1-year Committed Use Discounts.

Cevap

Deploy standard Compute Engine VM instances using a custom machine type configured with 10 vCPUs and 20 GB of memory.
Configuring a custom Compute Engine machine type allows exact sizing of 10 vCPUs and 20 GB RAM, preventing cost overruns from over-provisioned predefined machine types and software licenses. Using standard VM instances guarantees that the 4-hour job executes without preemption.

Adım Adım Çözüm

1
Analyze workload continuity requirements.
The batch job requires 4 hours of uninterrupted runtime and cannot tolerate preemptions.
Spot VMs may be terminated by GCP at any time with a 30-second notice, making them incompatible with non-fault-tolerant long-running tasks.
2
Analyze compute resource ratio and licensing impact.
The workload requires a non-standard 10 vCPUs to 20 GB RAM ratio.
Predefined machine types (such as standard, highmem, or highcpu families) force over-provisioning vCPUs or memory, increasing compute and third-party software licensing costs.
3
Select the optimal Compute Engine machine configuration.
Configuring a custom machine type with 10 vCPUs and 20 GB memory on standard VMs provides exact sizing without interruption risk.
Compute Engine custom machine types allow tailoring vCPU counts and memory capacity to fit exact workload demands.

Anahtar Kavram

Custom Machine Types and Spot VM Suitability
Bu soruyu puanla

Konular

Planning Compute Engine Resources and Machine Types

Bu konudan daha fazla soru