Benchmarking the Migration: Export and Import

This section covers the benchmarking of the export of data from the source system and the import of data into the target system as well. To carry out the benchmarking, SUM displays a sequence of dialogues, which you have to maintain.

Procedure

Get Roadmap

After you have chosen to work without a stack configuration file and selected the option Benchmarking Tool, choose the benchmarking mode Benchmark export and import and click Next.
Dialog Database Migration Option: Target Database: Select your target database from the list of the available target database types.
Dialog Migration Parameters
1. Table Selection: Select if the migration benchmark shall relate to all tables, that is, to the whole database, or to selected tables only.
  - If you choose All tables, you can specify additionally whether the benchmarking shall be performed for the whole database or on a sampling basis. Enter for this a value from 1 to 100 in percentage, where 100% means the whole database.
    
    If you enter a lower value, the system determines internally the size of all tables, sorts them according to the size, and selects the tables based on a sequence until the given size of the database is reached.
    
    In addition, you specify in percentage the maximum size of the largest table in the sample based on the database size.
    
    Example
    
    Assuming that the source database has a size of 1 TB and you want to export 10%. This means that 102.4 GB of the database is the sample. However, the number of tables depends on their size. If, for example, the database contains among others one table with 102 GB, the sample would contain only this table.
    Therefore, you specify the size of the largest table in the sample based on the size of the database. For example, 1% percent as maximum size means 1% of 1 TB database size. Hence, the largest table in the sample must not be bigger than 10.24 GB. In this case, the minimum number of tables in the sample is theoretically 10 tables with a size of 10.24 GB. The number increases with the number of smaller tables.
  - If you choose Selected tables, enter the name and the path of the txt-file that includes the table names. The txt-file contains each table in one line.
2. Migration Key: Provide the migration key for the data import.
3. Declustering during Migration: Activate the option Target database declustered, if you already know that the database will be declustered during the actual update and migration procedure, for example if you plan to update to SAP Kernel 7.40 or higher.
  
  This option allows you to receive the best possible simulation result.
4. Migration Repetition Option: Activate the option Enable migration repetition option to stop the benchmarking tool after the downtime migration phase and to enable a simple repetition of the latter to optimize the procedure by adjusting the parameters (such as R3load processes).
  
  Do not use this option for a productive run. For more information, see Using the Migration Repetition Option for Testing.
5. DMO with System Move: Activate Enable the migration with System Move if you want to move of the primary application server from source to target system.

Roadmap Step Configuration

Dialog Download Directory
1. Download Directory: Specify the download directory that contains, for example, the required kernel files so that the Software Update Manager can update the kernel automatically.
  - Note that the folder for the duration files must be existing, but not filled.
  - You can skip this step when only customer transports are imported.
2. Archive Verification: Activate the option Check the authenticity of SAP archives if you want to check the authenticity of all the archives in the download directory.
  
  For more information, see the chapter "Verifying SAP Archives" in section " Running the Software Update Manager" of the SUM guide.
Dialog Target Database Client: Specify the installation path to the database client. If the client is not yet installed, you specify the future path after the installation. In this case, you also have to specify the installation files in a subsequent dialog.
Dialog Database Migration Initialization: Specify the parameters for the migration initialization of the target database: hostname , SID , instance number.
Note
- The SUM displays in the dialog the SID of your source database. If the SID of the target database is different, adapt the SID accordingly.
- If your target database can only be reached via IP name and not via host name (for example for a cloned system), edit the etc/hosts file because the hdbsllib used for the connection resolves and uses the host name.
Dialog Required Migration-Specific Passwords. Provide the migration-specific passwords for different users on the target database, such as:
Example
- User SYSTEM
- User DBACOCKPIT (only if it does not yet exist on the target database and SUM has to create it)
- User SAP<SID> (Note: SUM creates it on the target database)
Dialog Schema Name in Target Database: Determine the database schema in the target database.
Dialog Parameters for Procedure: Configure the number of parallel SQL and R3load processes.

Note

For the benchmarking, Uptime means the Preparation roadmap steps, in which the tables for the export are determined, and Downtime means the Execution roadmap step, in which the table content is exported.

Roadmap Step Execution

During the roadmap step Execution, the benchmarking is executed.

Roadmap Step Postprocessing

Dialog Benchmarking completed: The procedure is finished. Analyze the utility DMO Migration Post Analysis in the SUM Utilities.

To access the SUM Utilities, open the More menu in the upper-right corner on the menu bar. For more information, see also SUM Utilities for DMO (Extended UI).
Fill in the feedback form that SUM displays.
Optional: Save migration duration files for performance optimization, as described in chapter Performance Optimization with Migration Duration File.
Clean up the procedure at the end. Afterwards you can start the SUM from the beginning.