Domain 5: Managing Implementation and Ensuring Solution and Operations Reliability — GCP PCA Practice Question 48

How to approach this question

Identify the managed Hadoop service, and the compute pricing model best suited for short, fault-tolerant batch jobs.

Full Answer

Cloud Dataproc Spot VMs (Preemptible VMs)

Cloud Dataproc (Option A) is Google's managed Hadoop and Spark service, allowing you to lift-and-shift existing Spark jobs easily. Because the jobs are batch processes running only 2 hours a night, they are highly fault-tolerant. Configuring the Dataproc cluster to use Spot VMs (Option C) for its worker nodes provides massive cost savings compared to on-demand pricing.

Common mistakes

Selecting CUDs (D). CUDs require a 1 or 3-year commitment for resources running 24/7. You would waste money on the 22 hours a day the cluster is off.

Your company is running a large Hadoop cluster on-premises and wants to migrate to Google Cloud. The workloads consist of ephemeral Spark jobs that run for 2 hours every night to process logs. Cost optimization is a primary concern. Which TWO GCP services/features should you use? (Select TWO)

How to approach this question

Full Answer

Common mistakes

Practice the full GCP Professional Cloud Architect Practice Exam 4

More questions from this exam