Best Practice for Initial Load

[former_member69564]

Hello,

what is the best way of doing the initial load? is there a best practice somwhere that tells you what should be imported first?

I want to understand the order ex,

1. load Lookups,

2. Hierarchies,

3. taxonomy and attributes

last the main table

etc...

I dont understand the logic.

Thanks in advance

[Former Member]

Hi Ario,

If you follow any SAP Standard business content for MDM Repositories like e.g. Material.

https://websmp130.sap-ag.de/sap/support/notes/1355137

In the SAP Note attachments, you will get MDM71_Material_Content.pdf

You will see Import of reference Data(look up table's data) 1st(step6) before import of Master data(step7).

During Import of Reference Data(look up data), Please follow the Import Sequence by using Processing level 0,1,2 etc.

Which take care of filling look up flat tables first then filling Hierarchies tables etc.

After that if you are maintaining Taxonomy, You need to fill taxonomy table in Taxonomy mode of Data Manager, in the sequence (Categories, Attributes, Linkage between Attributes and Categories and lastly Attribute Values)

After this I mean populating Reference data you need to populate Main table records along with tuples table data since now in MDM 7.1 Tuple has been replaced by Qualified table for most of the Master's but if you are still maintaining Qualified table you can import Mani table data along with Qualified table in a single step. Otherwise for Qualified table you can alos use this approach of populating Non-qualifeirs to Qualified table first before importing main table and then importing Main table data along with Qualifier's field of Qualified table.

This above entire process for exporting data from SAP R/3 system to MDM. If you are importing data into MDM from legacy system (Non-Sap systems too), Approach should be remain same Populating Lookup tables data and lastly main table data.

I dont understand the logic.

The logic is simple in your main table you have fields which are look up to Reference tables( e.g. field in main table which are look up to Lookup flat tables like Countries, Currencies etc, field in main table which is lookup to Hierarchy/Taxonomy table etc). So, if these values are not populated firstly, so during your Main table import you will have incomplete data for all of these fields from main table which are look up to some other tables as values in your lookup table you haven't populated before Main table import.

Kindly revert if you still have any doubts.

Regards,

Mandeep Saini

[Former Member]

Hi Ario

The sequence for initial load should be as follows-

1. load the lookup tables first. If you are using key maping for auto value mapping take care to map the remote keys as well.

2. Load the Hierarchy, Taxonomy and Attributes.

3. Load the Main table records. You can load data in Qualified tables (NQs) in the same go OR for loading the Qualified table you can also use a separate import map.

4. Save all the import maps with proper names which can be used for different environments.

5. Validate the data loaded by comparing the extracted file from MDM(using excel export in Data manager) and the source file to match the records cotent.

best regards

Ravi