Hard1 markMultiple Choice
Domain 2.2: Planning and configuring compute resourcesCompute EngineSpot VMsCost OptimizationLocal SSD

GCP ACE · Question 13 · Domain 2.2: Planning and configuring compute resources

You are designing an architecture for a nightly batch processing job. The job takes about 3 hours to complete, processes data from Cloud Storage, and is highly fault-tolerant (if a node fails, the master node simply reassigns the task to another node). You need to minimize compute costs.

Which TWO configurations should you include in your design? (Select TWO)

Answer options:

A.

Use Spot VMs (Preemptible VMs) for the worker nodes.

B.

Purchase a 3-year Committed Use Discount (CUD) for the worker nodes.

C.

Use Sole-tenant nodes.

D.

Store intermediate processing data on Local SSDs attached to the worker nodes.

E.

Store intermediate processing data in Cloud Storage Multi-Region.

How to approach this question

Identify cost-saving compute options for fault-tolerant workloads and performant storage for temporary data.

Full Answer

Use Spot VMs (Preemptible VMs) for the worker nodes., Store intermediate processing data on Local SSDs attached to the worker nodes.
For fault-tolerant batch jobs, Spot VMs (formerly Preemptible VMs) provide massive cost savings. Since the data is intermediate and temporary, Local SSDs provide the best performance/cost ratio for scratch space, as they are physically attached to the host server.

Common mistakes

Selecting CUDs for a workload that only runs 3 hours a day.

Practice the full GCP Associate Cloud Engineer Practice Exam 2

50 questions · hints · full answers · grading

More questions from this exam