MDM Staging and Loading 3 — Questions and Answers
Question 1: Which Informatica MDM configuration controls how many records are committed to the database in a single transaction during a load job?
- Trust settings
- Batch size (commit interval) (Correct answer)
- Column-level mappings
- Match threshold
Correct answer: Batch size (commit interval)
The commit interval (batch size) determines how many records are inserted or updated per database transaction, balancing throughput against memory usage.
Question 2: After running a stage job, a data steward notices the _STG table is populated but the base object remains empty. What is the most likely cause?
- The match job has not been executed yet
- The load job has not been executed yet (Correct answer)
- The cleanse functions rejected all records
- The XREF table has reached its row limit
Correct answer: The load job has not been executed yet
Staging and loading are separate jobs; populating the _STG table only completes the stage phase, and the load job must run separately to move records into the base object.
Question 3: What does the 'promote' operation accomplish within the Informatica MDM load process?
- It elevates a pending record to be the system-of-record winner
- It moves cleansed and validated staging records into the base object table (Correct answer)
- It applies trust decay rules to aging records
- It converts unmerged clusters into consolidated golden records
Correct answer: It moves cleansed and validated staging records into the base object table
Promotion is the step within the load job where records that passed cleansing and validation are written from staging into the base object, making them part of the hub.
Question 4: In Informatica MDM, which scenario would trigger a reject during the load job rather than the stage job?
- A record with a null required field from the source
- A record that violates a uniqueness constraint defined on the base object (Correct answer)
- A record with an unrecognized source system ID
- A record whose date field cannot be parsed
Correct answer: A record that violates a uniqueness constraint defined on the base object
Base object-level uniqueness constraint violations are detected at load time when MDM attempts to insert the record into the hub, not at the earlier stage validation phase.
Question 5: When an Informatica MDM landing table is used instead of a direct staging table approach, what is the landing table's role?
- It permanently stores golden records for downstream consumption
- It acts as a temporary holding area for raw inbound data before the stage job processes it (Correct answer)
- It holds match results awaiting manual review by data stewards
- It stores audit logs of all batch job executions
Correct answer: It acts as a temporary holding area for raw inbound data before the stage job processes it
Landing tables serve as the initial drop zone where source system data arrives, and the stage job then reads from landing tables to populate the formal staging tables.
Question 6: Which MDM Hub job is responsible for populating the _STG table from the landing table?
- Load job
- Stage job (Correct answer)
- Match job
- Automerge job
Correct answer: Stage job
The stage job reads data from landing (or external sources), applies cleanse functions, and writes results into the _STG staging table.
Question 7: In Informatica MDM, what is the significance of the LAST_UPDATE_DATE column in the staging table?
- It stores the date the record was first created in the source system
- It is used by delta detection to identify records modified since the last successful stage run (Correct answer)
- It records when a data steward last edited the record in the Hub Console
- It tracks the expiration date for time-limited records
Correct answer: It is used by delta detection to identify records modified since the last successful stage run
LAST_UPDATE_DATE is evaluated during delta detection to determine which source records have changed since the previous staging run, enabling efficient incremental processing.
Which Informatica MDM configuration controls how many records are committed to the database in a single transaction during a load job?