Basics of Mass Data Transfer

Before you begin the initial data load from a non-SAP system to an SAP system, you have to answer the following questions:

• Which SAP business object types are involved in the target system?

  You have to determine which business object type can be allocated the imported data in the destination system. This is important because the business dependencies are mapped across an object model in the SAP system. For example, if you want to transfer data involving sales processing, then business objects Material , Customer , and Sales Document are involved.

• How should the data be transferred to the SAP system?

  You have to determine the data transfer technique to use for each involved business object type. The techniques BAPI , batch input , and direct input are available for loading the data into the target system. For details of each technique, see the description of the Transferring the Data into the SAP System phase. The procedure for IDocs continues to offer an additional option for data transfer.

The key feature of the data transfer is that data can only be transferred from a non-SAP system to an SAP system when it is available in SAP format. For this reason the data is first to be extracted to a file from the non-SAP system and then put in a transfer file in SAP format. This transfer file is used as the basis for the actual dat aimport into the SAP system and can, for example, be created using mapping tools such as the Legacy System Migration Workbench (LSMW).

The entire data transfer usually takes place in five technical phases, which are illustrated in the following diagram.

These phases are described in detail below:

1. Analysis and cleanup of data in the non-SAP system

   This phase involves answering three related questions:

   • Which data exists in the non-SAP system?

   • How is this data structured (length, sequence)?

     Note

     This step is required to assign the transferred data to the correct SAP structures later.

   • Which data can be transferred unchanged, which data has to be supplemented, and which data cannot be transferred at all?

     Example

     A power company manages its customers for electricity, gas, and water in three separate systems. The company wants to merge these three systems in one SAP system. If customers exist who purchase both gas and water from this company, for example, then the data on these customers may have a different status in the different systems. For this reason, the customer data in each system first has to be cleaned up or merged before it can be transferred to the SAP system.

2. Extraction of data from the non-SAP system

   In this step, the determined data is extracted from the non-SAP system to a file. The data in this file still has the format of the non-SAP system. The following options are available for extracting the data:

   • Using an extraction tool supplied by the non-SAP system

   • Using the database tools

   • Using the extraction function of a mapping tool

3. Mapping the data in SAP format

   Data from a non-SAP system can only be transferred to an SAP system if it is available in SAP format. For this reason, the file containing the data in the non-SAP system format has to be transformed into a transfer file in SAP format.

   Because the format of the data extracted from the non-SAP system usually differs from the SAP format of the transfer file, mapping between the different structures is required. To perform this mapping, you will have to analyze both the fields and structures of the data to transfer and the involved SAP structures, in order to determine how the conversion should be performed.

   The following options are available for performing the actual mapping:

   • Using the LSM Workbench

   • Creating an ABAP program,

   • Using a commerical mapping tool

4. Transferring the data in SAP system

   Before you can import the data from the transfer file to the SAP system, you have to answer the following questions:

   • In which sequence will the source data be copied to the business object types, and in which sequence will the involved business object types be filled with data?

     This question must be answered because dependencies may exist between business object types that require the data to be transferred in a specific order.

   • Which data transfer technique will be used for the individual business object types?

     The following three standard techniques are available for the data transfer:

     • BAPIs

       Calling a BAPI in the appropriate application transfers the data to the SAP system. If BAPIs are used as interfaces to the SAP system, the same technique is used as for the continuous data transfer between SAP systems or between non-SAP systems and SAP systems with ALE.

       More information on data transfer using BAPIs: Process Flow of Mass Data Transfer Using BAPIs .

     • Batch input

       Batch input is a standard technique for transferring large volumes of data into the SAP System. The transaction flow is simulated and the data is transferred as if it were entered online. The advantage of this procedure is that all the transaction checks are performed, which guarantees data consistency.

       For more information on batch input, see the documentation on batch input sessions under Data Transfer: Overview of Batch Input .

     • Direct Input

       With direct input, data from the data transfer file is first checked to see if it meets various criteria and then transferred directly into the SAP system. The SAP database is updated directly with the transferred data.

5. Checking the data

   After the data has been transported into the SAP system, it must be checked for completeness. It must be ensured, for example, that the sum of the data transferred into the SAP system is the same as the sum of data from the external system. These checks are often made by check programs that are run in the SAP system. These programs can be supplied either by SAP or by an external provider.