Reviewed by Aditya Kumar · Last reviewed 2026-03-24
**Why throughput modes**: DynamoDB charges by read/write capacity. Provisioned = you set RCU/WCU; predictable cost. On-demand = pay per request; no capacity planning. **Provisioned**: Set RCU/WCU; pay per hour. Use for steady, predictable traffic. **Auto-scaling**: DynamoDB...
This easy-level Cloud/Tools question appears frequently in data engineering interviews at companies like Persistent Systems. While less common, it tests deeper understanding that distinguishes strong candidates.
Start by clearly defining the core concept being asked about. Interviewers want to see that you understand the fundamentals before diving into implementation details. Structure your answer with a definition, then explain the practical application with a concise example.
Why throughput modes: DynamoDB charges by read/write capacity. Provisioned = you set RCU/WCU; predictable cost. On-demand = pay per request; no capacity planning. Provisioned: Set RCU/WCU; pay per hour. Use for steady, predictable traffic. Auto-scaling: DynamoDB adjusts RCU/WCU based on target utilization (e.g., 70%). Min/max bounds. Use for variable traffic with known bounds. On-demand: Pay per request; scales automatically. Use for spiky, unpredictable traffic. Trade-off: Provisioned is cheaper at steady high load; on-demand is simpler and better for spiky. Auto-scaling bridges the gap—variable but bounded. Cost: At 1000 steady RCU, provisioned ~$150/month; on-demand at same load could be 2×. Best practice: Use on-demand for new/unknown workloads; switch to provisioned + auto-scaling when traffic stabilizes and you can profile.
Pro-Move: 'We use provisioned with auto-scaling for our main table—saved 40% vs on-demand; we use on-demand for staging tables with spiky load.' Red Flag: Using provisioned without auto-scaling for variable traffic—throttling or over-provisioning.
Some links below are affiliate links. If you buy through them we may earn a small commission at no extra cost to you — it helps keep DataEngPrep free.
According to DataEngPrep.tech, this is one of the most frequently asked Cloud/Tools interview questions, reported at 1 company. DataEngPrep.tech maintains an editor-reviewed database of 1,863 data engineering interview questions across 7 categories.