Post reply

ETL - DataWarehouse - Reporting

  • July 9, 2007 at 9:08 pm

    #177431

    For the whole time I work in data warehouse business, I learn from the ETL side, get the data and build the data warehouse with dimension tables and fact tables. In my company we use Cognos for reporting, it read the data warehouse dimension and fact tables to build the reports.

    If the user found out some problems in the report, I think I would go to the data warehouse side to find the problem. If the data warehouse data is fine, then I will question the Cognos reporting programmer.

    Recently something liked this happened, my new manager told me that I had to start with the ELT side to look for possible problems before asking the reporting manager, did it make sense?

  • July 10, 2007 at 3:14 am

    #718214

    Well, if the data in the datawarehouse is correct there is nothing problematic in ETL. I work for ETL and we just port the OLTP data into OLAP data fitting into the Warehouse schema. It works the same everywhere, so if the warehouse data is correct, ETL cannot be a problem origin.

  • July 19, 2007 at 11:25 pm

    #721006

    Whenever there is a bug in front end reports, the first thing people assume some thing wrong with the ETL or ETL has corrupted the data :-). My advice is to do root cause of analysis of the bug and identifying in which phase it has occured. Either problem with the source system, problem with the ETL etc..






    Regards,
    Sudheer 

    My Blog

Viewing 3 posts - 1 through 2 (of 2 total)

You must be logged in to reply to this topic. Login to reply