CompTIA Data+ (DA0-002)Data MiningMedium
A data architect is designing a new data ingestion pipeline that needs to process high volumes of streaming data from IoT devices. The data needs to be minimally transformed before being stored in a data lake for future analytical processing. The architect prioritizes speed and scalability for initial ingestion, deferring complex transformations until data is queried. Which data integration paradigm is best suited for this scenario?
- AData Mart
- BOLAP (Online Analytical Processing)
- CELT (Extract, Load, Transform)
- DETL (Extract, Transform, Load)
Show answer & explanationAnswer & explanation
Correct answer: C. ELT (Extract, Load, Transform)
ELT is ideal for scenarios involving high volumes of raw data, especially from streaming sources, where the data is first loaded into a scalable data lake (often cloud-based) with minimal or no transformation, and then transformed within the target system as needed. This defers computational burden and leverages the target system's processing power.
Why the other options are wrong
- A. A Data Mart is a subset of a data warehouse, a target for data, not a paradigm for data integration itself.
- B. OLAP is a type of analytical processing, not a data integration paradigm for ingesting raw data.
- D. ETL performs transformations *before* loading, which would slow down the ingestion of high-volume streaming data and add unnecessary processing overhead upfront.
ELT (Extract, Load, Transform)
A data integration process where data is first extracted from sources, then loaded directly into a target system (like a data lake or cloud data warehouse), and finally transformed within the target system as needed for analysis.
- Favored for large volumes of raw and unstructured data.
- Leverages the processing power of the target data warehouse/lake.
- Offers flexibility for schema-on-read and agile transformations.
Memory trick: Data flows in, gets changed, then finds its place.