Skip to main content

Legacy data load approach in BW/4HANA Shell Conversion

Published on 21 July 2021
BW/4HANA Shell Conversion
Sarath Kannan
Sarath Kannan
Senior Consultant and Practice Manager

Sarath Kannan is the Senior Consultant and Practice Manager of Business Analytics at Applexus. In this role, he manages the Analytics Practice, Consulting and Provides direction and guidance for operations, administration and results for the analytics team.

Introduction

SAP provides various options to migrate from legacy BW systems to BW/4 HANA:

  1. In Place: Convert the existing BW system to BW/4 HANA
  2. Remote: Install new system and selectively migrate BW Objects and Data (no selection option available for Data migration)
  3. Shell Conversion: Install new System and selectively migrate BW Objects. Conversion tool does not provide any options to migrate data; we will have to use standard BW techniques to load data.
BW/4HANA Shell Conversion
Source: SAP

Shell approach gives flexibility in changeover and helps fulfil the customer needs in a significant way with the salient features as mentioned below.

  • Supports the carve out scenarios more efficiently since there is no Requests association. This is unlikely with Remote or IN-place conversions
  • High flexibility for the customers in making decisions on the data retention policy of the legacy Data.
  • Advantage for the customer to build a clean BW/4HANA system with the option to cherry pick and carry forward their existing data flows with relevant data sets or even start fresh with the option to leave the legacy data behind. (Accelerated green field )
  • Flexibility to optimize the existing flows and incorporate the BW contents at the same time

Know more about shell conversion through BW/4HANA Webinar Series

In this blog we will be focusing on how to handle historical data loads during a BW/4 HANA migration using Shell Conversion. The 2 options available for loading legacy data are:

  • Load data from Source systems
  • Load data from legacy BW system

 Loading from original source system has the following disadvantages:

  • Not all the data will be available in the sources (retired legacy systems, data archived etc.)
  • Loading from original source does not allow loading data directly to the final targets in BW/4 and will have to go through the data flows requiring further troubleshooting.

Recommended approach: Load data from Legacy BW to BW/4 Data target using a 1:1 mapping

BW/4HANA

Legacy Data + Delta Management in Shell Conversion

BW/4HANA Shell Conversion

The above pic illustrates the high level dataflow into and within the BW/4HANA System from Original BW as well as from the Original source connected.

When we talk about the Historical data load using the ODP data sources from the Original BW and the relevant quiet time in original BW during the historical load; we can achieve this with a proper cutover plan.

Arrangements for Making B4H System for Historical Load

Source System & Data Sources

As an initial task to prepare the B4H box ready for the historical load we have to create a source system for the Original BW system in BW/4HANA.

BW/4HANA Shell Conversion

The Source system should be created as a Myself BW System (ODP_BW) in B4H which enables the developer to replicate the data sources corresponding to each info provider from the Original BW system and this can be used for further feeding historical data in B4HANA system.

Ref.SAP Note : https://launchpad.support.sap.com/#/notes/2481315

The Data sources are of a specific naming convention as below.

Data source Type

Naming Convention

Example

Master data Attributes

IOBJNAME_P

0MATERIAL_F

Master data Texts

IOBJNAME_T

0MATERIAL_T

Master data Hierarchies

IOBJNAME_H

0MATERIAL_H

Transaction Data Info providers

INFOCUBE / DSONAME_F

0IC_C03_F

Transformation & DTPs

We can create the transformations between the replicated data sources and the converted ADSOs /IOBJs through shell conversion. It will be a 1:1 mapping since both end fields are identical and the system itself will propose the mapping during the creation, however its worth to recheck the mapping and make sure it’s flawless.

We can create the DTP with relevant restrictions as per the data requirements and can trigger a full load either specific to the Info provider or through the process chains designed for historical data load.

The flow will be something similar as in the below pic once we created the Transformations & DTPs

BW/4HANA Shell Conversion

High Level Cutover Plan & Requirements

To successfully load historical data in a shell conversion path requires a proper cutover planning which includes the following check points

  • Make Sure all the Objects involved in the flow are Active in both the System ( Old /New BW )
  • Make sure all the delta records that have already been received from the source are uploaded in Original BW targets completely.
  • Both Original & New BW System should connected to same source to resume the Delta load later.
  • Stop all the process chains scheduled in the original BW and make sure no more data load to the Original BW until the release of quiet time.

General Reads

Add new comment

Plain text

  • No HTML tags allowed.
  • Lines and paragraphs break automatically.
  • Web page addresses and email addresses turn into links automatically.