Manage Batches and Pipelines DP-203 Practice Questions (Page 5)

Questions 21–25

1. 21application · medium

   A software company develops Azure Data Factory pipelines in a team environment. They want to implement version control for their pipeline artifacts so that multiple developers can work on different pipelines simultaneously and changes are tracked. They currently use Azure DevOps with Git repositories. What should they do?

   • AEnable 'Save as draft' in the Data Factory UI and rely on the built-in version history.
   • BExport all pipelines as ARM templates and store them in a separate Git repository.
   • CUse a shared Azure Storage account to store pipeline JSON files and manually copy them to the Data Factory.
   • DConfigure the Data Factory to use Azure DevOps Git repository and create feature branches for development.
   Reveal Solution Discussion Select an answer first
2. 22application · medium

   A data engineering team uses Azure Synapse Analytics to run Spark jobs that transform large datasets. They have a pipeline that triggers a Spark job definition every day. The Spark job sometimes fails due to out-of-memory errors when the input data size is larger than expected. The team wants to make the Spark job more resilient without manual intervention. What should they do?

   • AEnable autoscale on the Spark pool and set the dynamic allocation to true.
   • BAdd a retry policy to the pipeline activity that runs the Spark job.
   • CIncrease the executor memory in the Spark job definition to the maximum allowed.
   • DChange the Spark job to use a smaller file format like Parquet to reduce memory usage.
   Reveal Solution Discussion Select an answer first
4. 23application · medium

   A company uses Azure Data Factory to run a pipeline that processes orders from an ERP system. The pipeline must run every day at 2:00 AM, but on the first day of every month, it must also run at 6:00 AM to process the monthly close. The team wants to implement this with minimal maintenance. What should they do?

   • ACreate two separate pipelines: one for daily runs and one for the monthly run.
   • BUse a single schedule trigger with a recurrence of 'every day' and add a condition in the pipeline to check if it is the first of the month.
   • CCreate a single pipeline and attach two schedule triggers: one daily at 2:00 AM and one monthly on the 1st at 6:00 AM.
   • DUse a tumbling window trigger with a 1-day window and set the start time to 2:00 AM, then add a second trigger for the monthly run.
   Reveal Solution Discussion Select an answer first
5. 24foundation · easy

   A batch load in Azure Data Factory fails midway through a copy activity. What is the default behavior of the pipeline run?

   • AThe pipeline run is marked as Succeeded with warnings.
   • BThe pipeline continues with the next activity.
   • CThe pipeline run stops and is marked as Failed.
   • DThe pipeline automatically retries the entire pipeline from the beginning.
   Reveal Solution Discussion Select an answer first
6. 25foundation · easy

   When using Git integration in Azure Data Factory, what is the purpose of the 'publish' branch?

   • ATo store the published version of the Data Factory service
   • BTo store test data for pipelines
   • CTo store logs of pipeline runs
   • DTo store the working branch for development
   Reveal Solution Discussion Select an answer first