Reviewed by Aditya Kumar · Last reviewed 2026-03-24
**OLTP** (SAP, PostgreSQL): Optimized for high-volume, low-latency transactional writes; ACID; point lookups; indexes on primary/foreign keys. **OLAP** (Snowflake, Redshift): Optimized for bulk reads, aggregations, scans; columnar storage; few indexes. **Why separate**: OLTP...
OLTP (SAP, PostgreSQL): Optimized for high-volume, low-latency transactional writes; ACID; point lookups; indexes on primary/foreign keys. OLAP (Snowflake, Redshift): Optimized for bulk reads, aggregations, scans; columnar storage; few indexes. Why separate: OLTP can't scale for analytics without degrading transaction latency; OLAP schemas (Star) are write-inefficient. Adidas relevance: OLTP powers e-commerce checkout, POS, inventory—sub-100ms response. OLAP powers regional dashboards, demand forecasting, cohort analysis. Architecture: CDC (Debezium/Fivetran) streams OLTP → data lake → OLAP. Cost: Dedicated OLAP prevents OLTP overprovisioning; analytics runs on burst compute.
Red Flag: Running analytical queries directly on OLTP. Pro-Move: 'We moved all BI queries off the transactional DB; OLTP P99 latency dropped from 800ms to 45ms, and we avoided a $2M hardware refresh.'
According to DataEngPrep.tech, this is one of the most frequently asked SQL interview questions, reported at 1 company. DataEngPrep.tech maintains an editor-reviewed database of 1,863 data engineering interview questions across 7 categories.