IBM C1000-190: IBM watsonx Data Lakehouse Engineer v1 - Associate

Last checked November 10, 2026 against the official IBM exam page.

1. Question 1Data Lakehouse Fundamentals

   A retail company stores clickstream logs as raw JSON files in object storage. Analysts want to run SQL on them, and data scientists want to train models from the same tables. Which architecture best meets this need?

   1. AKeep separate copies in a traditional warehouse and a data lake, and synchronise them by hand every night using scheduled manual scripts that someone must monitor
   2. BLoad all logs into one relational server and scale it vertically as the volume grows over time
   3. CUse an open table format on object storage with a shared metadata layer that multiple engines can query
   4. DCopy the logs into a second warehouse product, so each team works on its own isolated extract of the data
   Show answer and explanationHide answer

   Correct answer: C

   A lakehouse uses an open table format with shared metadata over object storage, so SQL engines and data science tools read the same governed tables. Manual nightly copies create drift and extra cost, a single relational server does not suit raw file volumes at scale, and spreadsheets cannot serve SQL or model-training workloads.

2. Question 2watsonx.data Fundamentals

   A data team has registered an Iceberg catalog in watsonx.data and wants analysts to query its tables with a Presto engine. What must be true for the engine to see those tables?

   1. AThe catalog's data files must first be copied onto the local disk of every engine worker node in the cluster
   2. BThe engine must be associated with the catalog so that it can resolve table metadata and storage locations
   3. CAnalysts must export each table to CSV files that the engine then reads through a separate batch job
   4. DThe engine must be restarted after every query so that it refreshes the catalog listing for the requesting user
   Show answer and explanationHide answer

   Correct answer: B

   In watsonx.data an engine can only use the catalogs it is associated with, which gives it access to table metadata and storage locations. Copying files to workers, exporting CSVs or restarting engines for each query are not required to query catalog tables.

3. Question 3Operation

   A data steward wants an analyst to see only the non-sensitive columns of a customer table, while a compliance officer can see every column. Which control is most suitable?

   1. AOne shared administrator login for all users so that every person sees the same consistent view of the data
   2. BA copy of the table with the sensitive columns deleted and no further access controls applied to that copy
   3. CColumn-level access policies that grant each role access only to the columns it is permitted to see
   4. DFull table access for every user, with reviews of the query history once each month after the data is used
   Show answer and explanationHide answer

   Correct answer: C

   Column-level policies enforce different visibility per role on the same governed table. A stripped copy creates a second table to maintain, a shared login removes accountability, and monthly history reviews do not prevent exposure.

4. Question 4Data Integration

   A source system sends new and updated customer rows every hour. The lakehouse table must reflect the changes without reloading the full history on each run. Which ingestion pattern fits best?

   1. AAn incremental load that merges only new and changed rows into the target table on each scheduled run
   2. BA full truncate-and-reload of the entire table from the source system every single hour of the day
   3. CA manual export of the source table to a spreadsheet at the end of each week for the analysts to review
   4. DDropping the target table and recreating it from the schema of the first load on every scheduled run
   Show answer and explanationHide answer

   Correct answer: A

   Incremental loads with a merge step capture only new and changed rows, which keeps hourly runs fast and the target current. Full reloads, weekly manual exports and drop-and-recreate jobs either reload unnecessary history or fall behind the source.

5. Question 5Consumption

   An analyst filters a partitioned lakehouse table on its partition column. The engine reads only the partitions that can match the filter. What is this behaviour called?

   1. AColumn projection, which limits the number of columns that are returned to the client application
   2. BPredicate broadcasting, which copies the filter condition to every table that exists in the catalog at once
   3. CPartition pruning, which skips partitions that cannot satisfy the filter condition of the query
   4. DQuery federation, which merges the results of several separate engines into one final result set for the user
   Show answer and explanationHide answer

   Correct answer: C

   Partition pruning uses the filter to skip partitions that cannot contain matching rows, which reduces the data read. Column projection, federation and predicate broadcasting are different mechanisms and do not describe skipping partitions.