Loading Table Structure FAQS

1. What is a loading table?
• Answer: A loading table is a temporary table created in the staging area during the loading process. It is typically prefixed with C$ and holds data that is required for further transformations or integration in the staging area.
2. How is a loading table different from a source datastore?
• Answer: A loading table represents an execution unit and does not directly map to a source datastore. Unlike a source datastore, which contains all of the data from the source system, the loading table only contains the data that is needed for transformations or integration within the staging area.
3. Why are only specific attributes included in the loading table?
• Answer: Only the attributes required for transformations or integration in the mapping are included in the loading table. This ensures that unnecessary data is excluded, improving performance and making the data processing more efficient.
4. How does the filtering of data work in the loading table?
• Answer: If a filter is applied to the source data (e.g., filtering by CUST_AGE in the CUSTOMER table), the loading table will only contain the filtered records. Attributes not needed for transformations or integration, such as CUST_AGE, will not appear in the loading table.
5. What happens if I join multiple tables in a mapping?
• Answer: If two or more tables are joined in a mapping (e.g., CUSTOMER and SALES_REPS), the loading table will contain the combined attributes from both tables. The data from the joined tables will be included in the execution unit for further processing.
6. Will the loading table always be an exact copy of the source datastore?
• Answer: No, the loading table is not always an exact copy of the source datastore. If only certain attributes or data from the source are required, the loading table will only contain those attributes. However, if no transformations or joins are applied and the entire source datastore is mapped, the loading table will be an exact image of the source datastore.
7. Why do we need to use loading tables in the staging area?
• Answer: Loading tables in the staging area are used to temporarily store data that is necessary for transformations and integration processes. They help organize data before it moves to the final target system, allowing for clean processing and transformations in the staging area.
8. Can loading tables be used for sources that don't support transformations, like flat files?
• Answer: Yes, even for sources that don't support transformations, like flat files, the loading table can be created to hold the entire source data. The loading table will then act as an exact copy of the source datastore, and any necessary processing will occur within the staging area.
9. Do loading tables affect the performance of the mapping process?
• Answer: While loading tables are useful for organizing and processing data, the performance of the mapping process can be affected by the amount of data being loaded. Limiting the data to only necessary attributes and applying filters at the source can help improve performance.
10. What happens to the loading table after the integration phase?
• Answer: After the integration phase, the loading table is typically dropped to clean up the staging area and free up resources. This ensures that no leftover data from previous loading processes affects future mappings.