Try OpenEdge Now
skip to main content
Database Administration
Maintaining and Monitoring Your Database : Table-partitioned database : Partitioning your database
 

Partitioning your database

With a high-level understanding of table partitioning, you can begin the process of partitioning your database. While a specific process is unique to every situation, the following is a high-level process outline:
*Understanding what tables you will partition.
You will need to evaluate the layout and content of your database to determine the best tables to partition and how they are partitioned (list, range, sub-partition). Consult OpenEdge Getting Started: Table Partitioning for a discussion of the factors you should consider when planning your partitioning.
*Understanding your physical layout.
You will need to evaluate the current storage layout of your database. As you partition your database, do you need new storage areas to hold your new partitions? Do you want to move tables and indexes from their existing storage areas? Use the existing PROSTRCT and PROUTIL tools to make these changes.
*Enabling partitioning.
Prior to creating partitions, you will need to enable your database. You can use PROUTIL or OpenEdge Management/Explorer to enable your database.
*Defining partitions
Once you have a plan, you have several options to define your partitions:
*OpenEdge Management/Explorer provides a graphical interface to define partitions. See Getting Started: OpenEdge Management and OpenEdge Explorer and OpenEdge Explorer: Managing Table Partitioning in Databases.
*OpenEdge SQL provides DDL for partitioning. See OpenEdge Data Management: SQL Development and OpenEdge Data Management: SQL Reference for details.
*OpenEdge ABL provides APIs for partitioning. See OpenEdge Development: Programming Interfaces for details.
*Migrating data
When you first create partitions on an existing table, the data remains in the composite initial partition. You can use the PROUTIL PARTITIONMANAGE SPLIT option to move data out of the initial partition and into other partitions. If you have a new partitioned table, loading data through any of the standard load options (command line or tools) the data is automatically populated into the correct partitions.