none
Request for Incremental (cube/partition) processing approach

    Question

  • Hi,

    Can you please help me on to do the incremental partition processing.

    Example.

    I have some records in the fact table. I add some more records into the fact table. now I need to process the update data only instead of deleting all records and insert the data again and process.

    Thanks,

    Narendra

    Wednesday, July 11, 2018 12:05 PM

All replies

  • you can use ProcessAdd operation for that

    see

    https://www.mssqltips.com/sqlservertip/3073/sql-server-analysis-services-processadd-for-a-partition/


    Please Mark This As Answer if it solved your issue
    Please Vote This As Helpful if it helps to solve your issue
    Visakh
    ----------------------------
    My Wiki User Page
    My MSDN Page
    My Personal Blog
    My Facebook Page

    Wednesday, July 11, 2018 12:14 PM
  • Hi Narendra,

    As Visakh mentioned, you can use process add for the partition to only bring in new records.

    If you are using tabular cubes then please follow these links.#

    In general - https://www.sqlbi.com/articles/incremental-processing-in-tabular-using-process-add/

    for model compatability > 1200 - https://www.sqlbi.com/articles/using-process-add-in-tabular-models/

    HTH,

    Ram


    Please vote as helpful or mark as answer, if it helps


    Wednesday, July 11, 2018 1:15 PM
  • Hi Narendra,

    Thanks for your question.

    Visakh16 and Ram have already provided solid solutions for this issue. Please kindly mark the helpful replies as answers. By doing so, it will benefit all community members who are facing this similar issue. Your contribution is highly appreciated.

    One thing I want to add, please take care of the SQL Query you are using for partition, you need to make sure that the same records has not been imported multiple times. otherwise you might encounter the issue in below thread:
    Replicate values in OLAP cube


    Best Regards
    Willson Yuan
    MSDN Community Support
    Please remember to click "Mark as Answer" the responses that resolved your issue, and to click "Unmark as Answer" if not. This can be beneficial to other community members reading this thread. If you have any compliments or complaints to MSDN Support, feel free to contact MSDNFSF@microsoft.com

    Thursday, July 12, 2018 1:52 AM
    Moderator
  • You can "roll" your update partition forward by using views for each partition and some some dateadd(getdate()) logic in the view where clause.  If it's a multidimensional model you'll also need to update the ssas slice property of the partition using some .NET code in an SSIS script component in order to keep it synced with the view where clause.  This helps you keep a very small refresh partition without maintenance.

    I have used this in conjunction with Proactive Caching to achieve near real-time cube loads.
    Thursday, July 12, 2018 2:05 AM