Domain 4: Data Integration

IBM Certified watsonx Data Lakehouse Engineer v1 - Associate · 55 questions

• Which method would be most appropriate for ingesting a large historical dataset from an on-premises relational database into watsonx.data?
• What does data virtualization in watsonx.data enable that traditional ETL does not?
• To query a PostgreSQL database from Presto in watsonx.data without moving data; what must be configured?
• Which external databases can be connected to IBM watsonx.data for data virtualization through JDBC connectors?
• How does IBM DataStage integrate with watsonx.data for data pipeline workflows?
• How would you ingest real-time event streams from Apache Kafka into watsonx.data?
• What PySpark API would you use to write a transformed DataFrame as an Iceberg table in watsonx.data?
• When ingesting JSON files with unknown structure into watsonx.data using Spark; what does schema inference do?
• When connecting an external PostgreSQL database to watsonx.data; what determines how PostgreSQL data types are mapped to Presto types?
• When creating an Iceberg table in watsonx.data to store daily transaction data; what partitioning strategy would most improve query performance for date-range queries?
• What is the primary use case for IBM Data Replication in the context of watsonx.data?
• When running a federated query that joins a large Iceberg table in watsonx.data with an external PostgreSQL table; what technique improves performance?
• After a streaming Spark job writes thousands of small Parquet files to an Iceberg table in watsonx.data; what operation should be run periodically?
• When configuring a JDBC connection to an external database in watsonx.data; why is setting appropriate connection pool limits important?
• When ingesting data into watsonx.data for analytics; which target file format is recommended for new tables to maximize query performance?
• What Spark code pattern is used to read data from an external JDBC database into a DataFrame for ingestion into watsonx.data?
• What is an advantage of using watermark-based incremental loading over full table refreshes when ingesting from a source database into watsonx.data?
• Which Spark Structured Streaming write mode would you use to continuously append new Kafka events to an Iceberg table?
• For a use case where a data scientist needs to run occasional ad-hoc queries joining watsonx.data tables with a live IBM Db2 operational database; which integration approach is most appropriate?
• A Presto query filters on WHERE eventdate = '2025-01-15' on an Iceberg table partitioned by eventdate. What performance optimization occurs?
• A reporting team needs to join lakehouse order tables against a customer table that lives in an operational PostgreSQL database. The customer table changes constantly and the join runs a few times a day. What is the appropriate approach in watsonx.data?
• An analyst has a notebook that cleans and reshapes raw ingest data, and three teams now depend on running it themselves before reporting. What should change?
• An organisation needs to bring data from a transactional database, a SaaS application and an existing data warehouse into its lakehouse. What is the correct approach?
• A new ingestion pipeline is being designed for a source that feeds regulatory reporting. Which requirement should the design satisfy?
• A team federates a legacy reporting database that takes about ninety seconds to answer a typical aggregate query. They are surprised the federated version is no faster. Why?
• A finance team maintains a monthly revenue restatement by exporting a lakehouse table to a spreadsheet, applying adjustments by hand, and uploading the result. What is the principal risk?
• A source system is queried infrequently, is subject to strict residency rules, and its owners will not permit copies. Which integration pattern fits?
• What is the consequence of loading data without validating its structure?
• A team argues that federating a source is a way to give analysts access to data that governance would otherwise restrict, because the data is never stored in the lakehouse. How should this be assessed?
• A transformation job completes with no SQL errors, but downstream users report that a revenue column is now null for most rows. What does this indicate about the pipeline's design?
• A dashboard must reflect operational data within a few seconds of it changing. What does this requirement imply for the integration design?
• Why should metadata be captured as part of ingestion rather than added once the pipeline reaches production?
• An architect proposes copying every federated source into the lakehouse first, then federating those copies, to guarantee consistent performance. What is wrong with this plan?
• What most clearly distinguishes a controlled transformation pipeline from ad hoc transformation work?
• Why is federation an unsuitable choice for a workload that repeatedly runs heavy aggregations over a large source table?
• A team proposes ingesting only the aggregated totals their current dashboard shows, rather than the underlying detail. What risk does this create?
• Which requirement most strongly indicates that federation is the right choice rather than ingestion?
• A pipeline produces a table that several reporting teams consume. What should accompany the table so consumers can use it correctly?
• A newly configured connector authenticates successfully but returns no rows for a table the source team confirms has data. What should be checked first?
• A daily load occasionally contains records with an unexpected extra field. Which handling approach best preserves both data quality and continuity?
• A federated query against a remote warehouse is slow, and monitoring shows a very large volume of data being read across the network for a query whose result is a few hundred rows. What should be examined first?
• A team plans to run transformation logic directly against production tables, overwriting them in place with each run. What is the main objection?
• What is the primary factor in deciding between ingesting a source into the lakehouse and federating queries to it?
• Which combination describes a production-ready ingestion pattern?
• A team plans to query three systems - an operational database, an existing warehouse, and lakehouse tables - in a single report. What does federation in watsonx.data provide here?
• Two teams report different totals for the same metric. Investigation shows each maintains its own transformation deriving the metric from the same raw tables. What is the appropriate fix?
• A team wants to replace their brittle collection of hand-run export scripts with a supportable integration into watsonx.data. What should they adopt?
• Data quality problems in a governed table are consistently discovered by business users reading reports. What does this indicate about the ingestion design?
• After six months, a federated connection to a partner's reporting database is generating enough load that the partner asks the team to reduce it. The queries are dashboard refreshes over largely historical data. What is the right change?
• A pipeline's source system adds two columns and renames a third. The pipeline continues to run successfully but the renamed column's data disappears from the output. What practice would have caught this before consumers noticed?
• Which risk is introduced by transferring source data as manually prepared files rather than through a supported connector?
• Which statement about ingestion into a governed lakehouse is correct?
• Which statement about federation in watsonx.data is accurate?
• Why is testing a transformation important even when the SQL executes without error against production data?
• Which statement about connecting disparate sources to watsonx.data is correct?