Writing to cube slow ?!

This site uses cookies. By continuing to browse this site, you are agreeing to our Cookie Policy.

  • Writing to cube slow ?!

    We are running all Jedox services on the same server which uses SSDs only. Although the process of writing data into the cube seems to be quite slow:

    Source Code

    1. 2017-10-11 15:13:57 INFO Retrieving data from transform TR_Norm_Data
    2. 2017-10-11 15:14:23 INFO Records loaded to Cube: 100000
    3. 2017-10-11 15:14:48 INFO Records loaded to Cube: 200000
    4. 2017-10-11 15:15:15 INFO Records loaded to Cube: 300000
    5. ...
    6. 2017-10-11 15:18:10 INFO Records loaded to Cube: 1000000
    7. 2017-10-11 15:18:36 INFO Records loaded to Cube: 1100000
    8. ...

    So about 4 Minutes for 1 m records. This consumes quite an amount of time you have a total number like

    Source Code

    1. Lines normalized from transform TR_Norm_Data: 21695344

    Do you have similar experiences concerning cube load speed?
    Any idea how I can test where a potential break could originate?
  • jjunek wrote:

    What version are you using? Which process uses the cpu most? Are you splashing or loading base cells?

    7.0 SR2
    During Cube Load: mainly palo.exe, but also tomcat.exe
    Splashing is disabled, mode is "update"
    Not sure what you mean with "loading base cells"

    laloune wrote:

    Any SupervisionServer workers on the cells you are populated?

    Where can I check this? SVS starts by default with standard worker, i.e. palo.ini:

    Source Code

    1. worker "E:\Jedox Suite\svs\SupervisionServer.exe"

    The post was edited 1 time, last by holgrich ().