DEV EPM

Now its time to glue it all together with ODI. ODI is great here because it can work with different technologies without any effort. In our case, we will run the Java codes that we displayed in this series using Java BeanShell Technology.

Although ODI is great to execute any kind of technology code out there, it does not have a good way for you to write and debug your code while you are developing it. So, I always prefer to create the code in an external Java IDE (like Eclipse), test it and copy the “Main” portion of it into an ODI Procedure.

Another advantage to use ODI is that we can get the connection information from topology itself and reuse everywhere within the package using the command on source/target technique, where you define the connection in the command on source tab and get the information in the command on target tab.

To glue all together in ODI is very simple. We may have all the java codes in one procedure with as many steps as we want (depending in what kind of statistics we want to get, in our case, three). Then we may have another procedure that will hold our pivot queries that we use to transform and load the data into our DW tables. Finally, we may even create our own metrics based on the knowledge that we have from Essbase. Down below is one example of metrics that we may retrieve from the stats that we just gathered.

When we put it all in an ODI package, it will look like this. In this example, we also added a send email component just to inform the users about the job completion.

That’s it folks! Next post will be the last one of this series. Stay tuned!

This entry was posted on February 19, 2021 at 3:51 pm and is filed under Essbase, Java, ODI with tags Essbase, Java, ODI. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.