You can add a Load profiling process to a Quality workflow to run a metadata analysis and load data into a dynamic entity. The Load process is used on dynamic entities and output from Quality processes. The process creates one real entity, from which you drill down to view results that include:
- Analyzed attributes
- Analyzed business rules, the resulting rules compliance percentage, and the total number of passing and failing business rules
- Discovered keys, dependencies, and joins
- The maximum and minimum data row lengths
- Data rows loaded into the output entity
Additional attribute metadata is available, including:
- Total count and distribution percentage of values, nulls, spaces, distinct values, and strings among others
- Metaphones, soundexes, masks, and patterns
Note: Running the Load process on a dynamic entity can be used in place of the Regenerate
TSQ Data option.
Guidelines
- Use the Load process if your focus is to load data into a dynamic entity and analyze the data or to re-load and re-analyze data in a real output entity. After you run the Load process, you can add other profiling processes for further data analysis.
- See example:
-
- The Load process can only be used on a dynamic input entity or on the output from a Quality process; it cannot be used on an entity that is already fully loaded. For example, if the input entity is real, add a Transformer process first, then add the Load process.
- Running the Load process discovers potential keys and dependencies on a sample of data from the input entity. If you want to instead create dependencies and keys by specifying specific attributes, use the Keys and Dependencies processes.
- If there are no results for dependencies or keys analysis, then none were discovered.
- Before you can add an Analysis, Business Rules, Dependencies, or Keys process to a Load process, you must first run the Load process.