Hitachi Vantara Pentaho Community Forums
Results 1 to 3 of 3

Thread: Pentaho BI/Carte master/slave cluster and questions

  1. #1

    Default Pentaho BI/Carte master/slave cluster and questions

    Hey all,
    I'm reading up on the clustering support for Kettle (2.5.X series at the moment) and I just would like confirmation on what I understand (from an ignorance standpoint, I'm still catching up). If there are changes happening for 3.0 for some of these, please go ahead and list seperately:

    *This will only work with Carte servers. I have not found a reference to using a Pentaho BI server as a master, but this scenario just may not have been discussed.

    *Will only work with transformations, period. As I understand, if you have a transformation in a job, you can NOT run the job and have the transformation run clustered. Please let me know otherwise.

    *Clustering will help processing large number of rowsets and/or time-consuming steps/transformations even if you do not take advantage of Database Partitioning.

    *Other than the Carte documentation, there is no other resources regarding clustering (if so, please list them as I'm relying on posts in the forums and mailing lists).

    *EDIT: Does using carte have certain repository requirements?

    thanks,
    -D
    Last edited by dhartford; 08-14-2007 at 10:33 AM.

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

    Default

    *This will only work with Carte servers. I have not found a reference to using a Pentaho BI server as a master, but this scenario just may not have been discussed.
    ok

    *Will only work with transformations, period. As I understand, if you have a transformation in a job, you can NOT run the job and have the transformation run clustered. Please let me know otherwise.
    For now yes... job support was supposed to be in 2.5.0 but was moved

    *Clustering will help processing large number of rowsets and/or time-consuming steps/transformations even if you do not take advantage of Database Partitioning.
    k

    *Other than the Carte documentation, there is no other resources regarding clustering (if so, please list them as I'm relying on posts in the forums and mailing lists).
    Not that I know of for the moment

    *EDIT: Does using carte have certain repository requirements?
    It's more push for the moment, than pull.

  3. #3
    Join Date
    Nov 1999
    Posts
    9,729

    Default

    No repository requirements. What happens is that you design a transformation T to run in clustered mode on several carte slave servers.
    T then gets split into parts T-Master and T-Slave-1 through T-Slave-N.
    The posting of these derived transformations to the (carte) slave servers happens using XML only with HTTP servlets.

    HTH,
    Matt

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.