DB2 extenders can operate with DB2 Extended Enterprise Edition, and in this way take advantage of the partitioned database support provided by DB2 Extended Enterprise Edition.
A partitioned database is a database that is distributed across two or more independent machines. To the end-user and application developer, the database appears as a single database on a single machine. Partitioning allows applications to efficiently use a database that is simply too large to be handled by one machine.
A partitioned database is composed of two or more partitions. Each partition is managed by its own database partition server. A database partition server includes a database manager and the collection of data and system resources that it manages. Typically, one database partition server is assigned to each machine. However, it is possible to have multiple database partition servers on a single machine. Each database partition server holds a portion of the entire database. A database partition server is also sometimes called a node.
As Figure 9 illustrates, database partitions can be grouped logically and assigned a name. Each group of database partitions is known as a nodegroup. Defining nodegroups allows you, for example, to limit application queries to selected database partitions, and thereby speed up transaction times. A nodegroup can contain one database partition only, or it can contain multiple database partitions. If a nodegroup contains multiple database partitions, it is known as a multipartition nodegroup. All database partitions named to a multipartition nodegroup must reside within the same database.
Figure 9. Nodegroups in a database
![]() |
Using the extenders in a partitioned database system means you can: