configuring and ODS

[Former Member]

Hi guys,

I am trying to figure out what situations i can use an ODs instead of a cube, and how do i select what items are data and what items are keys inside the ODS? Thanks in advance.

[Former Member · Accepted solution]

Here are the Scenarios for DSO.

1) SAP recommands DSO to be used in the first layer. It is called Enterprise DataWarehouse layer. Though, not all business contents come with this DSO layer, you may want to build your own.

2) There is always the need for faster data load. DSOs can be configured for Write optimized. Thus, the data load happens faster and your load window is shorter.

3) Datasouce is nor delta enabled. in this case, you would want to have an ODS first and pull Delta to a cube.

4) If Multi Dimensional analysis not required and Flat Table-like structure suffice. In this case, you might want to use DSOs and can avoid overheads.

5) If your data needs to be overwritten or updated. Cubes does not support Overwite or Delete directly.

<u>How to determine the Keys :</u>

<i>Analyze the data and identify the keys that should identify unique records. The way you select the keys and update mode (Overwrite/Update/ingnore duplicate records) makes a big difference what is finally stored in your DSO.</i>

Regards.

<b>If this satisfies your question, Awarding points is a way of thanking !</b>

[Former Member]

Key fields, as opposed to data fields, define unique records in an ODS; they are the "primary fields" in the database. Usually, you should only define a number of key fields necessary so that unique records are not overwritten.

ODS's can be useful when you need to change certain records that are already uploaded to BW (i.e. delta), because in an ODS, you can change existing records, whereas in a cube you can't (in this case you'd need to send new records to the cube that "cancel" the existing records, for example)