Implementing Siebel Business Applications on DB2 UDB for z/OS > Migrating Data Using Siebel EIM > About Setting Up EIM for DB2 >

Choosing an Appropriate Partitioning Scheme

To achieve the best system throughput, partition base tables and EIM tables and set up parallel EIM processes to efficiently exploit partitioned EIM table spaces.

When a table is created on a partitioned table space, the table is assigned a partitioning key composed of one or more columns. EIM tables by default use the clustering U1 index, IF_ROW_BATCH_NUM, ROW_ID. In some cases, you can obtain better performance by partitioning EIM tables based on the partitioning key of the target base table, for example, on IF_ROW_BATCH_NUM plus the partitioning key of the corresponding target table.

NOTE:  Do not update the values in the partitioning key because you might have to perform an unload and reload or a REORG on the data to resolve performance issues.

For more information on DB2 partitioning, see Understanding Siebel Partitioning.