Wolters Kluwer
Contact: Don McGimpsey
Country: USA
Award Category: Innovative Problem Solver
Tell us about the business problem you were trying to solve.
I used the export from an Oracle data map to create a metadata mapping table. The metadata contains all column names, and attributes. From this table, the user can chose their own mapping of source to destination fields, and rollup, indexes, sort columns and even expressions (Combining address fields to create consolidated address for instance). This metadata is used to dynamically write SQL and execute it to create an ETL job, and we use this to pull data from over 600 tables into a total of 12 new tables. The end user only needs to edit an excel file and save as CSV.
What SAS products did you use and how did you use them?
What were the results or outcomes?
Time savings of great significance. This prototype has been adapted over and over again for other ETL tasks, and is very reliable.
Why is this approach innovative?
It separates the end user from the raw ETL process and leaves the data processing management to an administrator.