Data Management & Integration Flashcards

A retailer's MDM hub must merge customer records from five source systems while leaving each source as the system of record. Which MDM implementation style fits best?

Answer: Registry style

Registry style keeps data in the sources and maintains only a cross-reference index with matching keys.

Which technique captures only inserted, updated, and deleted rows from a source database's transaction log for downstream integration?

Answer: Log-based change data capture

Log-based CDC reads the database transaction log, so it picks up every change with little load on the source.

In a data mesh, which principle most directly addresses the bottleneck of a single central data engineering team?

Answer: Domain-oriented decentralized ownership

Domain ownership moves accountability for analytical data to the business domains that produce it.

A Type 2 slowly changing dimension handles a customer's address change by doing what?

Answer: Inserting a new row with effective dates and a current flag

SCD Type 2 keeps full history by adding a new versioned row with effective dates.

Which DAMA-DMBOK knowledge area covers the movement and consolidation of data within and between data stores and applications?

Answer: Data Integration and Interoperability

DMBOK's Data Integration and Interoperability area covers ETL, ELT, replication, messaging, and virtualization.

A CDO needs real-time query access across three databases without physically copying the data. Which approach fits best?

Answer: Data virtualization / federation

Data virtualization gives one logical view by querying sources in place at run time.

Why does an ELT pattern often suit cloud data warehouses better than traditional ETL?

Answer: It uses the warehouse's scalable compute to transform data after loading

ELT loads raw data first and then transforms it inside the warehouse, using elastic cloud compute.