DEA-C01 Practice Question 55: Redshift Serverless Sporadic Workload
AWS Certified Data Engineer – Associate (DEA-C01) · Test 6 · question 5 of 10
A finance team's Amazon Redshift workload is concentrated into the first three business days of each month, when a dozen analysts run heavy SQL. For the rest of the month the provisioned cluster is almost entirely idle but still billed. The team wants the same Redshift SQL, the same data-lake queries through Redshift Spectrum, and a bill that reflects actual usage, without anyone pausing and resuming a cluster on a calendar.
Which option should the data engineer recommend?
Show answer and explanation
Answer:
Redshift Serverless "automatically provisions data warehouse capacity and intelligently scales the underlying resources", letting you "access and analyze data without the need to set up, tune, and manage Amazon Redshift provisioned clusters" and "pay only when the data warehouse is in use." That is the whole ask: the three busy days are billed, the idle weeks are not, and nobody manages a schedule. Cost control is explicit too: "you can set a base capacity in Redshift Processing Units (RPUs)… You can also set usage limits, such as RPUs used per day, to control costs." The SQL surface is unchanged, and data-lake queries are supported, so Spectrum-style querying across S3 keeps working.
Option A adds capacity at the peak but leaves the baseline cluster running and billable for the other 27 days, the actual source of the waste.
Option C does address idle cost, but it is the calendar-driven operations the team explicitly rejected. Pause/resume is coarse, an unexpected mid-month query finds the cluster paused, and the Lambda schedule becomes one more thing to maintain and get wrong.
Option D changes the engine and the capability set. Athena cannot join to Redshift-managed tables, loses Redshift-specific SQL, workload management and materialized views, and forces a full migration of the warehouse for what is fundamentally a billing-shape problem.
Reference: What is Amazon Redshift Serverless?