Skip to Content

Welcome to the Data Federator (DF) blog.

Data Federator is part of the Information Management (IM) family but its multi-source capabilities have a nice fit with our BI offering.
Data Federator has been designed to solve the multi-source query execution problem and it is a fantastic complement for BOBJ Universes or Data Services.

Data Federator has mainly 2 components
– DF Designer
– DF Query Server

DF Designer
This component is used to design a data model (aka. Data foundation), register the sources you want to federate and define the mappings between the sources and the data model (target tables).

DF Query Server
This component behaves like a virtual databases.
At runtime, the DF Query Server receives a SQL query, optimizes it and rewrites the SQL query into a collection of sub-queries send to the underlying sources.

I will come back on these components and their features in the coming months.

To report this post you need to login first.

18 Comments

You must be Logged on to comment or reply to a post.

  1. Gregory Blair
    Frederic,

    I’m certainly intrigued by Data Federator.  What are the best uses for Data Federator?  I haven’t run into many people using it and want to know if they’re really missing out on something great or if they know something I don’t know.

    For example, would Data Federator be a good choice to serve up data to a call center dashboard.  There’s a great need in my company (Blue Cross Blue Shield of North Carolina) for a call center dashboard but we don’t have the budget for a project to do it right.  Would Data Federator be a good choice to pull the information together from 5 or 6 sources to feed a central Call Center dashboard?  Thanks!

    (0) 
    1. Frederic VANBORRE Post author
      Greg,

      Data Federator is able to query your data from where they live. You don’t have to move them ahead of time.

      The product has a few key unique concepts
      – A data modeling tool helping you to create your own data model
      – A mapping mechanism helping you to map your sources to your data model
      – A SQL engine coupled with a SQL optimizer to generate native statements which can be delegated to the underlying sources optimally.

      DF can be used to connect to your sources, create a data model which suits your dashboard needs and query in real time the underlying systems.

      In addition, DF connects to relational and non relational systems helping your query systems it was difficult to reconciliate before.

      Drop me an email if you need more details about DF and we will discuss if it is a good match for your requirements.

      Hope this helps,

      Fred

      (0) 
      1. Gregory Blair
        Frederic,

        I’d love to get some more details about DF.  One of the big concerns is the product roadmap.  One of my team members attended a Webinar on XI 3.0 and the person mentioned some developmental efforts.  The one that caught my eye was Multisource Universe.  It sounds like at least some DF functionality might be incorporated into Universe Designer.  Am I reading that correctly?  Thanks for your help.

        -Greg

        (0) 
      2. Gregory Blair
        I’d love to get some more details about DF.  One of the big concerns is the product roadmap.  One of my team members attended a Webinar on XI 3.0 recently and the presenter mentioned some developmental efforts that are underway.  The one that caught my eye was Multisource Universe.  It sounds like at least some DF functionality might be incorporated into Universe Designer.  Am I reading that correctly?

        What I’m trying to avoid is suggesting that we invest in DF only to find that it gets incorporated into a product we already own (e.g., Designer).  I’m also worried that I just don’t hear a lot of people talking about it.  It doesn’t seem to be used by very many people.  Do you have any examples of customers that are using it successfully?  Thanks for your help.
        (0) 
  2. Chapman Brad
    I created a Data Federator project that connects to 20 servers across US and Canada. All data sources are SQL Server 2005. The DF project maps 40 source objects into 4 target objects.  When one of the servers is offline (power outage) we get an error opening a Webi report referencing the DF connection. 

    This seems brittle and I suspect I am just missing some key information.  We should be able to connect to all available sources and at the same time query an object that can indicate any failed connections in the DF project.  Can you shed any light regarding how to handle connections or connector properties in order to address data sources that are not available?

    (0) 
    1. Frederic VANBORRE Post author
      Hi

      Could you contact me by email (firstname . lastname at sap dot com) to discuss about your deployment?

      Data Federator provides a few mechanisms like the intelligent data connection for instance but I need more details before to move forward with a solution.

      TIA

      Frederic Vanborre

      (0) 
  3. Srikanth P
    Hello Fred,

    I have a question regarding the usage of SQL Server stored procedures inside data federator. We have many SQL stored procedures that have business rules and we would like to directly access them in DF by mapping the stored proc output to target table fields. Can we do that? If so, could you please explain the steps?

    (0) 
    1. Frederic VANBORRE Post author
      Hi Hendrik

      Yes, DF leverages BWA for InfoCubes.
      BW transparently delegates indexed InfoCubes queries to BWA so the DF response times are extremely fast.

      Hope this helps,

      Fred

      (0) 
  4. jyoti jain
    I would like to know that my BO XI 3.1 server is installed in different machine.

    And my Data Federator server XI 3.0 SP2 is installed in a different machine and in that machine BO client is also installed.

    1) Do I require DF client also to be installed  ?

    2) Is it possible to install the DF server in a different  machine, where my BO server is not installed ? What parameters do i need to set so that Df projects can be used by Universes ?

    2) How would I be able to use the DF projects in my universe as a connection?  What parameters do I need to change? DO i have to … I have already referred the PDfs of Data Federator…but there is no such information in it…So can you please guide me on an urg basis and reply me ASAP

    (0) 
    1. Frederic VANBORRE Post author
      Hi

      1) Do I require DF client also to be installed ?

      Yes, you have to install the Data Federator client driver (aka thindriver.jar) on the BOE machine connecting to DF.

      2) Is it possible to install the DF server in a different machine, where my BO server is not installed ? What parameters do i need to set so that Df projects can be used by Universes ?

      Yes, this is the recommmended approach.

      Have a look to this post
      Your 1st DF project in less than 15min (part 2)

      In the connection definition, replace localhost by the hostname of the machine running the DF Query Server.

      2) How would I be able to use the DF projects in my universe as a connection?

      Cf. my previous answer.

      What parameters do I need to change?

      Replace localhost by the DF QS hostname.

      Hope this helps,

      Fred

      (0) 
      1. jyoti jain
        Thanks a lot for the Prompt response Fred.

        But just to clarify one thing,

        1) Do I have to do the DF installer (i.e client)full installation
        OR
        2) Only my changing the jdfc.so file in BO Enterprise Server will work to establish universe the connection?

        (0) 
        1. Frederic VANBORRE Post author
          It is recommended to do the DF client installation or to copy the thindriver.jar on the BOE machines.

          Once the thindriver.jar is copied on the BOE machines don’t forget to update the jdbc.sbo to reference it as explained in the data access guide.

          Basically, updating the ClassPath variable of the jdbc.sbo file to point to the thindriver.jar in the Data Federator section of the file.

          Dunno if it will work in the blog comment but you should end up with something like

          C:\\Program Files\\Business Objects\\BusinessObjects Data Federator 12 Drivers\\JdbcDriver\\lib\\thindriver.jar

          Hope this helps,

          Fred

          (0) 
  5. jyoti jain
    I would like to know the following things.

    1) We have installed Data Federator 3.0 SP2 , FP2 would it be compatible with BO XI 3.1 SP1

    2) If we installed the Data Federator in different machine and my BO server is installled in a different machine. Is it a right approach to do so ? If yes/ No, with reasons if possible?

    (0) 

Leave a Reply