reyemsaibot

SAP BI Blog about SAP BW/4HANA, Analysis for Office and SAP HANA - by Tobias Meyer

Create a parent-child hierarchy in SAP Data Warehouse Cloud

Update 01/2025

Name change from SAP Data Warehouse Cloud to SAP Datasphere. Some links may break

There are different ways to create hierarchies in Datasphere. One way is to use a CSV file and upload it into DSP. Another way is to use the existing hierarchies of your SAP BW system. In this
post, I want to show how to use a hierarchy from BW and transform it into a parent-child hierarchy in Datasphere. After the transformation, we use the hierarchy in a view. 

 

First, we need a hierarchy on one InfoObject in SAP BW. 

SAP BW InfoObject with hierarchy
SAP BW InfoObject with hierarchy

As we have now the hierarchy on the InfoObject, we can now transfer the H-Table as a remote table into Datasphere. For the transfer, we use the ABAP Connection and select under ABAP Tables the
corresponding table. With click on next step, we can import and deploy the table.

Import remote table into Datasphere
Import remote table into Datasphere

Now we have the H-Table in the Datasphere and can build an SQL View with a parent-child hierarchy. So let’s create a new SQL View in the Data Builder of the Datasphere.

Create new SQL View in Datasphere
Create new SQL View in Datasphere

I created a SQL View. The statement select the node name as parent id and the node name as child id from my hierarchy table.

 

 select ( select "NODENAME" 
            from  "SPACE./BIC/HZNETWORK" as a 
           where a."NODEID" = b."PARENTID" ) as PARENTID,
        ( select "NODENAME"
            from "SPACE./BIC/HZNETWORK" as a 
           where a."NODEID" = b."NODEID" ) as CHILD_ID
 
   from "SPACE./BIC/HZNETWORK" as b

 

The result looks like this

Data preview in Datasphere
Data preview in Datasphere

As you see, the hierarchy node 1 has a child with the id 2. The BW view looks like this:

Hierarchy in SAP BW
Hierarchy in SAP BW

Now we can use this view for an association with the dimension Sales Network (The InfoObject is ZNETWORK) (see figure 1) and use the dimension in the Sales View for filtering and navigation (see
figure 2).

Figure 1: Association hierarchy and dimension
Figure 1: Association hierarchy and dimension
Figure 2: Association dimension and view
Figure 2: Association dimension and view

In the SAP Analytics Cloud (SAC), the data looks like this:

Output of example in SAC
Output of example in SAC

Conclusion

This is just an example how you can create a parent-child hierarchy. SAP will offer such an option later when you look on the roadmap. I think this is an idea and if anyone has a better solution,
please feel free to share it. One restriction has the parent-child hierarchy at the moment. There are no text nodes supported yet. Maybe it comes with the next releases of Datasphere. Feel free
to share your thoughts on this idea in the comments.

author.


Hi,

I am Tobias, I write this blog since 2014, you can find me on Twitter,
LinkedInFacebook and YouTube. I work as a Senior Business Warehouse Consultant. In 2016, I wrote the first edition of . If you want, you can leave me a PayPal coffee donation. You can also contact me directly if you want.




Kommentare

4 Kommentare zu „Create a parent-child hierarchy in SAP Data Warehouse Cloud“

  1. Avatar von Sebastian Gesiarz
    Sebastian Gesiarz

    Thank you! You saved me a ton of work 🙂

  2. Avatar von Tobias
    Tobias

    Hi, you are welcome. Good to hear that it helps someone 😉

  3. Avatar von Rajat
    Rajat

    Here is easy way but the SQL:SELECT b.“NODENAME“ as „CHILD_ID“, c.“NODENAME“ as „PAREANT_ID“ FROM „HV0DFJ5G“ AS a INNER JOIN „HV0DFJ5G“ AS b ON a.“NODEID“ = b.“PARENTID“ INNER JOIN „HV0DFJ5G“ AS c ON a.“NODEID“ = c.“NODEID“ WHERE a.“HIEID“ = ‚I1MCNOLFGIDW3H25IPSEBY6KV‘ and b.“HIEID“ = ‚I1MCNOLFGIDW3H25IPSEBY6KV‘ and c.“HIEID“ = ‚I1MCNOLFGIDW3H25IPSEBY6KV‘

  4. Avatar von Tobias
    Tobias

    Hi Rajat, you are right. There are several ways to archieve this goal. Depending on the knowledge of SQL it is most of the time easier to use SQL instead of the graphical view.

Schreibe einen Kommentar

Deine E-Mail-Adresse wird nicht veröffentlicht. Erforderliche Felder sind mit * markiert