Imagine you have an ADO.NET Entity Framework that maps entities over an IBM IDS Datasource (DB2 or Informix).
Now imagine that someone asks you to replicate that data in an MSSQL database transforming it to exactly match the EF object schemas ( table and column names, column data-types, etc ). Consider this in large scale as well. Consider that this
database may house a few hundred tables, and 500,000 to 1,000,000+ rows per table.
Given the work that has already gone into defining the EF, it would be nice if I could at least stub out an ETL package based on the EF that is already defined.
That would be preferred over duplicating the effort in SSIS. Worse, having to iterate over every entity, and every object instance in each entity. Those options seem very slow, and very inefficient. Hoping there is a better, smarter, accelerated
way?