Oracle Transactional Business Intelligence

Welcome to the Oracle Analytics Community: Please complete your User Profile and upload your Profile Picture

BIP compress output prior delivery

649
Views
9
Comments

Summary

Compress output prior delivery

Content

To divide the large data set in multiple data set in OTBI we have option to use bursting.
By using bursting we have written logic which is working , but when we are compressing that output at the time of scheduling, system is putting same file in different zip.
We are using ( compress output prior delivery) option.

Example :
In Busrting Test XLSX format zip  we're running REP_SPC_DEM_INV_VIS_CUBE1 BIP report and splitting report output in multiple chunks with delivery channel used as SFTP.
Following zip folders are created at SFTP path,
REP_SPC_DEM_INV_VIS_CUBE_0
REP_SPC_DEM_INV_VIS_CUBE_1
REP_SPC_DEM_INV_VIS_CUBE_2
REP_SPC_DEM_INV_VIS_CUBE_3

But if you open all these folder, you can see it contains same file i.e. REP_SPC_DEM_INV_VIS_CUBE_1.xlsx.

So our ask is, why system is putting same csx or exls file in different zip folder? is it bug or product limitation?

Comments

  • rajArun
    rajArun Rank 5 - Community Champion

    and all the files have the same data?

  • FPonte
    FPonte Rank 6 - Analytics Lead

    Hi Niketa.

    On your bursting query. How is PARAMETER2 (File Name) defined?

    Cheers

    Fernando

  • Niketa Ashwini
    Niketa Ashwini Rank 2 - Community Beginner

      'Itronpdhdev' PARAMETER2,

  • FPonte
    FPonte Rank 6 - Analytics Lead

    I'm really sorry Niketa.

    I asked about the wrong parameter in the bursting definition.

    I mean to ask PARAMETER5 (Remote File Name) as you are sending the data to FTP right?

    Cheers

    Fernando

  • Niketa Ashwini
    Niketa Ashwini Rank 2 - Community Beginner

    This works fine when "Compress output prior to delivery" option is unchecked at the time report job submission.

    But when we select "Compress output prior to delivery", system splits CSV and created multiple zip files but puts same CSV is all zip, which is incorrect.

  • Niketa Ashwini
    Niketa Ashwini Rank 2 - Community Beginner

    Reports returning millions of record and size is also in MB's. so to Split CSV file we use bursting with del channel "FTP" and ORA_HASH function.

     

  • Niketa Ashwini
    Niketa Ashwini Rank 2 - Community Beginner

     

      'REP_SPC_DEM_INV_VIS_CUBE_' ||TO_CHAR(SYSDATE, 'DDMMYYHH24MISS')
      ||hash
      ||'.csv' PARAMETER5,
     
  • FPonte
    FPonte Rank 6 - Analytics Lead

    Hi Niketa.

    Thanks for your reply.

    Based on your previous comments that the process works fine when you don't zip, I think you have a bug in your hands (unless someone else has another view).

    It will be good to open a SR for Oracle Support.

    Cheers

    Fernando

  • Robert Fighiroae-Oracle
    Robert Fighiroae-Oracle Rank 4 - Community Specialist

    Hi Niketa,

    The "Compress output prior to delivery" option will zip the output of the whole report scheduled output and not at individual chunks that are delivered separately via the bursting logic.

    The customer should check the XML Data Chunking option for extremely large data sets.

    The XML data chunking capability in BI Publisher reduces the time and memory used for processing large data of size greater than 2 GB.  The concepts are similar to bursting where a single large XML data will be split into several smaller chunks of XML data of manageable size, be executed as multiple sub-jobs in parallel to process the individual XML data chunks, and finally merges the outputs to create a single consolidated output.

    Check below link for details and limitations:

    https://www.oracle.com/webfolder/technetwork/tutorials/tutorial/cloud/r13/wn/common/releases/19C/19C-common-wn.htm

    In the end, the customer could try using XML Data Chunking + Compress output prior to delivery, as the output will be a single file.

    Best regards,

    Robert