ETL versus ELT, data lakes, warehouses, star and snowflake schemas, Airflow, dbt, partitioning and data quality.
40 cards · basic cards · AI-written, checked twice. Edit anything.
- What does ETL stand for?
Show answer
Extract, Transform, Load- What does ELT stand for?
Show answer
Extract, Load, Transform- In ETL, where does data transformation occur?
Show answer
Between extraction and loading, in a separate transformation stage- In ELT, where does data transformation occur?
Show answer
After data is loaded into the target system- What is the main advantage of ELT over ETL?
Show answer
Data is loaded faster, and transformation leverages the target system's native processing power and SQL- When should you use ETL instead of ELT?
Show answer
When data must be cleaned and validated before loading, or when the target system lacks transformation capabilities- What is a data warehouse?
Show answer
A centralized repository that stores processed, structured historical data from multiple sources for analytics and reporting- What schema approach does a data warehouse typically use?
Show answer
Schema-on-write: the schema is defined and enforced before data is loaded- Why are data warehouses typically denormalized?
Show answer
To reduce the number of joins and optimize read performance for analytical queries- What types of queries are data warehouses optimized for?
Show answer
Complex analytical queries (OLAP) that aggregate large amounts of historical data across many rows and columns- What is a data lake?
Show answer
A large centralized repository that stores raw, unprocessed data in its native format without predefined schema- What schema approach do data lakes use?
Show answer
Schema-on-read: the structure is applied and interpreted when data is accessed, not when stored- What is the main difference between a data lake and a data warehouse?
Show answer
A data lake stores raw, unstructured data with flexible schema; a warehouse stores processed, structured data with fixed schema- What types of data can be stored in a data lake?
Show answer
Any type: structured, semi-structured (JSON, logs), and unstructured (images, videos, text) in raw form- What is a star schema?
Show answer
A dimensional model with one central fact table connected directly to multiple denormalized dimension tables
FlashKeepers