Migrate an on-prem Apache Spark 3 batch job (executors: 8 vCPU, 16 GB) to GCP with minimal changes, ability to choose similar resources, and minimal management overhead; job reads from Cloud Storage and writes to BigQuery.
Choose an answer
Tap an option to check your answer.
Correct answer: Run the job with Dataproc Serverless..
Why this is the answer
Dataproc Serverless is the best choice because it allows you to run Spark jobs without managing clusters. It automatically provisions compute resources (including vCPU and memory, similar to the on-prem setup) and scales them based on workload, minimizing operational overhead. It natively integrates with Cloud Storage and BigQuery, simplifying data access. Running on a new GKE cluster would require managing Kubernetes, increasing overhead. A Compute Engine VM would necessitate manual Spark installation and cluster management. A new Dataproc cluster, while offering managed Spark, still involves cluster provisioning and scaling decisions, which Dataproc Serverless abstracts away.
Pass your exam — without the endless answer hunt
Get every verified question and explanation for this exam in one place, and save hours of prep. 1,000+ certifications · 20+ languages · free to start.
Pass your exam faster → No card needed