Hitachi Vantara Pentaho Community Forums
Results 1 to 4 of 4

Thread: Synchronize

  1. #1

    Question Synchronize

    I don't know how to synchronize some very simple activities as inserting rows in two different dipendent tables.
    I don't want to use two transformations scheduled by a job!
    When I try to write a single transformation with two "subtask", I do not know how to ensure that the second "subtask" start when the first is finished.

  2. #2
    Join Date
    May 2006
    Posts
    4,882

    Default

    If you want to do it in 1 translation your only option I can think of is to put a blocking step (which passes all rows) in between. The blocking step will only release the rows when all rows have been processed before the blocking step. If you run using subtasks blocking step won't help as the subtasks will start without waiting for row input.

    Steps run concurrently and there's not much you can change about it. Why wouldn't you run 1 job running 2 transformations, that would definitely work.

    Blocking step passing all rows is available as of v2.5

    Regards,
    Sven
    Last edited by sboden; 07-12-2007 at 02:51 PM.

  3. #3

    Default

    Thank you very much, for your quick response.

    I do not want to use jobs because they hide the details of each transformation. Sometimes it's good, but often it isn't.
    For example, I want to see all the "subtasks" that write some correlated tables in the same transformation. I do not want to manage 10-15 different files to do a simple inserting operation.

    Why is it so difficult to add synchronization?
    I know that all steps run concurrently, but why is there no way for a step to wait for the end of the execution of another step?
    Something like the "stream lookup".

    Regards,
    Marco

  4. #4
    Join Date
    May 2006
    Posts
    4,882

    Default

    Why is it so difficult to add synchronization?
    It's currently not in the "model".... mostly performance reasons. You have jobs to sequentialize stuff, transformations to do the processing.

    I know that all steps run concurrently, but why is there no way for a step to wait for the end of the execution of another step?
    Something like the "stream lookup".
    Blocking step... to synchronize between steps (like insert/update) that should do the trick

    Regards,
    Sven

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Privacy Policy | Legal Notices | Safe Harbor Privacy Policy

Copyright © 2005 - 2019 Hitachi Vantara Corporation. All Rights Reserved.