IBM Sterling Transformation Extender

Sterling Transformation Extender

Come for answers, stay for best practices. All we're missing is you.


#Sterling
#Supplychain
 View Only
  • 1.  MPI_FILE_PREFIX_UNIQUE constant missing from dtxpi.jar

    Posted 09/06/12 11:01 PM

    Originally posted by: ten8ciousb


    WTX 8.4.0.0 (141)
    developing interface using the RMI API
    using these map settings

    map.setIntegerProperty(MConstants.MPIP_MAP_WORKSPACE_LOCATION, 0, MConstants.MPI_LOCATION_FILE);
    map.setIntegerProperty(MConstants.MPIP_MAP_WORKSPACE_FILE_PREFIX, 0, MConstants.MPI_FILE_PREFIX_UNIQUE);

    If I include/reference the dtxpi.jar, when compiling the java code I get this error
    MPI_FILE_PREFIX_UNIQUE cannot be resolved or is not a field

    If I include/reference the dstxpi.jar, when compiling the java code I do not get this error.

    If search the documentation and tried internet searches. My first thought was that MPI_FILE_PREFIX_UNIQUE was deprecated, but the 8.4 documentation still lists it.
    Questions:
    1. Is this a WTX bug?
    1.a. Is it resolved in later 8.4 releases? (I have not been able to find it in fix lists)
    2. Or is it a documentation discrepancy?

    I am using FILE instead of MEMORY for the workspace location partly because I'm updating an old Datastage interface.
    Broader Questions:
    1. Is there an advantage/disadvantage to using FILE over MEMORY? I haven't found documentation that explains this.
    2. is there an advantage/disadvantage to using PREFIX_UNIQUE over PREFIX_MAP?
    #IBM-Websphere-Transformation-Extender
    #IBMSterlingTransformationExtender
    #DataExchange


  • 2.  Re: MPI_FILE_PREFIX_UNIQUE constant missing from dtxpi.jar

    Posted 09/20/12 12:44 PM

    Originally posted by: repanzer1


    In reference to using work files or memory, the documentation states;

    When the size of the input data a map is processing is large, file-based workspace using default paging is recommended (Workspace = File).

    When small amounts of data are processed, smaller page sizes and memory-based workspace is recommended (Workspace = Memory).

    In the event you're processing small amounts of data, it's quicker if the system does not have to keep accessing files, but if you're processing large amounts of data, the system will have to use a lot of memory to compensate, and it's a safer bet to use files.

    Can't help with any of the other questions, unfortunately.
    There are 10 types of people in this world. Those who understand binary and those who don't.
    #IBMSterlingTransformationExtender
    #IBM-Websphere-Transformation-Extender
    #DataExchange


  • 3.  Re: MPI_FILE_PREFIX_UNIQUE constant missing from dtxpi.jar

    Posted 09/20/12 01:08 PM

    Originally posted by: ten8ciousb


    thank you @repanzer1 for taking the time to reply.
    I understand the reason why would would choose MEMORY over FILE or vice versa. I really didn't ask the question correctly. I should have asked at what point does it make sense to use FILE? What constitutes large amounts of data? 500MB? 5GB?

    As for my problem of the missing constant. Somehow at somepoint I had copied a dtxpi.jar file from an 8.2.0.3 version of WTX into my CLASSPATH.
    The constants are missing from the 8.2.0.3 version of dtxpi.jar, not the 8.4.0.0 version. Which I'm guessing is a bug, but obviously resolved in the 8.4 version. I I don't know if there is an 8.2.0.x version.
    #IBMSterlingTransformationExtender
    #DataExchange
    #IBM-Websphere-Transformation-Extender