Pattern Ideas for Distributed data Mining?


EJB design: Pattern Ideas for Distributed data Mining?

  1. Pattern Ideas for Distributed data Mining? (1 messages)

    I'm hoping to brainstorm for a possible project I may be helping with: I need a reporting system that does reporting on multiple databases in remote locations. Now, I know this is vague, and i've thought of a good number of different ways to do this, but i'm wondering if there is an established pattern for this type of thing. Since the databases are remote, I'd definatly like to cache as much as I can on the reporting system, but I'll have to have a good system for flushing the cache when modifications are made to the original database. Any ideas or links to patterns would be really appreciated. Thanks guys.
  2. Hi,
    Don't know if this is of much help, but I have heard that Oracle supports sendinfg JMS messages from triggers. You could use an asynchronuous suystem in order to update your database reoplica, but this means that you may have stale data sometimes. However, having stale data for a reporting system seems reasonable enough.
    Best regards, Mircea