1. Which of the following statements about the Window recipe is true?

In order for a Window recipe to work, all three Window definitions (Partitioning columns, Order columns, and Window frame) need to be activated.

In order to correctly compute the rank for each row, an Order column must be specified.

On the Aggregations step, you can compute custom aggregations with the DSS formula language.

If you define multiple windows, each window must be computed with the same Partitioning columns.

2. Which of the following statements is true of metrics and checks?

Metrics and checks are always recomputed after a dataset is rebuilt.

Metrics and checks are always used in conjunction with a scenario.

You can compute metrics and checks for each partition of a partitioned dataset.

Metrics and checks are found on the Statistics tab of a dataset.

Metrics can be computed for datasets, recipes, models, and managed folders.

3. Use the Advanced Designer Certificate Assessment project to answer this question. Why did the schema consistency check fail after updating the Prepare recipe?

The expected number of rows changed due to the removal of TEST rows.

An expected column used in a recipe downstream was removed.

The storage location of the output was changed.

The values in a column are different from what was expected after changing a Formula.

4. Imagine a Flow with more than 100 datasets and recipes. Which views are most likely to provide useful information for identifying the parts that take a long time to compute? Select two choices.

The Connections and Recipe engines views to see if connections and engines can be optimized for heavy-computation areas.

The Creation view to see when objects were created.

The Count of records view to identify the largest datasets in the Flow.

The Last build view to see which datasets were built recently.

5. Which of the following statements about the Pivot recipe is false?

Columns with over 20 categories cannot be selected as pivot columns.

Possible aggregations include: count, max, sum and average.

The schema of the output dataset cannot be computed until the first time a Pivot recipe is run.

Multiple columns can be defined as row identifiers.

If no row identifier columns are selected, there is a single row of output.

6. A smart recursive build is a safe way to ensure schema changes have been propagated throughout the Flow.

You need to use the schema propagation tool instead.

You need a forced recursive build if there are schema changes.

Yes, this will take care of it with the minimal amount of computation.

A simple non-recursive build will do it!