Reviewed by Aditya Kumar · Last reviewed 2026-03-24
CDC captures inserts, updates, and deletes from a source and applies them to a target in near real-time, enabling minimal-downtime migrations. **Approaches**: Log-based CDC (Debezium, AWS DMS)—reads WAL/redo logs; lowest latency, no schema change. Trigger-based—triggers on...
Red Flag: Claiming trigger-based CDC is 'real-time' without acknowledging write amplification. Pro-Move: Mention handling schema evolution (e.g., Debezium SMT) and idempotent writes to avoid duplicates during retries.
This easy-level System Design/Architecture question appears frequently in data engineering interviews at companies like Moonfare, Snowflake. 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.
CDC captures inserts, updates, and deletes from a source and applies them to a target in near real-time, enabling minimal-downtime migrations. Approaches: Log-based CDC (Debezium, AWS DMS)—reads WAL/redo logs; lowest latency, no schema change. Trigger-based—triggers on source; adds load and schema coupling. Timestamp/version columns—incremental only; misses deletes and out-of-order updates. Dual-write with reconciliation—applications write to both; eventual consistency and complexity. Why log-based: Non-invasive, captures all changes, low source overhead. Scalability: Kafka as CDC backbone allows multiple consumers and backpressure handling. Cost: DMS/MongoDB Atlas CDC have per-hour costs; Debezium is OSS but requires Kafka infra. Trade-off: Initial full snapshot + CDC is required; plan for schema evolution and idempotent upserts.
Red Flag: Claiming trigger-based CDC is 'real-time' without acknowledging write amplification. Pro-Move: Mention handling schema evolution (e.g., Debezium SMT) and idempotent writes to avoid duplicates during retries.
Master 179 system design/architecture questions with expert answers. Real questions from 97+ companies.
22 min read →Kafka is in every data engineering job description, but most candidates only know 'producers and consumers.' Master these 15 questions covering partitioning strategy, exactly-once semantics, and Kafka Connect patterns.
16 min read →Interviewers don't ask 'build a pipeline.' They ask 'how would you handle late data, schema changes, and exactly-once processing?' Master the 7 patterns that answer these questions.
15 min read →Compare your response with a stronger, structured example and identify missing concepts.
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 System Design/Architecture interview questions, reported at 2 companies. DataEngPrep.tech maintains an editor-reviewed database of 1,863 data engineering interview questions across 7 categories.