ClusterGather
We assume that you have already learned what is described in:
Short Summary
ClusterGather gathers data records from multiple cluster nodes.
| Component | Same input metadata | Sorted inputs | Inputs | Outputs | Java | CTL |
|---|---|---|---|---|---|---|
| ClusterGather | yes | no | 1-n (virtual) | 1 | - | - |
Abstract
ClusterGather receives potentially unsorted data records through single input port. ClusterGather gathers records as specified in the Merge type attribute.
Icon
Ports
| Port type | Number | Required | Description | Metadata |
|---|---|---|---|---|
| Input | 0 | yes | For input data records | Any |
| Output | 0 | yes | For gathered data records | Input 01) |
Legend:
1): Metadata can be propagated through this component.
ClusterGather Attributes
| Attribute | Req | Description | Possible values |
|---|---|---|---|
| Basic | |||
| Merging type | The way how records should be gathered. By round robin, sorted by key or gathered as sonn as possible. | Simple gather (default) | Round robin | Merge by key | |
| Merge key | yes | Key according to which the sorted records are merged if Merge by key is selected. (Remember that all key fields must be sorted in ascending order!) See Group Key for more information.1) | |
| Equal NULL | By default, records with null values of key fields are
considered to be different. If set to true,
they are considered to be equal. | false (default) | true | |
| Data policy | Determines what should be done when an error occurs. See Data Policy for more information. | Strict (default) | Controlled | Lenient | |