Distribution Partition

Note: this is an abstract component, that cannot be instantiated.

The Distribution Partition defines a partition for a distribution algorithm.

A distribution algorithm splits a subtree into several partitions. Each partition has an identifier which is unique throughout all the topology of servers. This means that when two distributions are configured on two servers and they are referring to the same data partition, they must refer to that partition using the same partition identifier. This is a requirement because indexes can be replicated between servers. Each partition has its own parameters. The partition parameters depend on the type of the partition. For example, partition parameters for a numeric distribution consist of a lower boundary and upper boundary. Each partition is mapped to a workflow element. The workflow element mapped to the partition can be of any type (load balancer, final workflow element) but from the partition perspective the workflow element is a data container.

Direct Subcomponents

The following Distribution Partitions are available in the server :

• Capacity Distribution Partition
• Dnpattern Distribution Partition
• Lexico Distribution Partition
• Numeric Distribution Partition

These Distribution Partitions inherit from the properties described below.

Relations From this Component

The following components have a direct AGGREGATION relation FROM Distribution Partitions :

• Workflow Element

Relations To this Component

The following components have a direct COMPOSITION relation TO Distribution Partitions :

• Distribution Algorithm

Properties

A description of each property follows.

Basic Properties

partition-id

workflow-element