User Guide

10.2 HotFix 1
- 10.5 HotFix 2
- 10.5 HotFix 1
- 10.5
- 10.2
- 10.1
- 10.0 HotFix 1
- 10.0

Back Next

Multi-Clustering-Definition

Use multi-clustering definition to define parameters for a cascade of clusters. The Data Clustering Engine clustering process uses the multi-clustering definition to run the cascade of clusters. The clustering process starts with the multi-clustering definition keyword.

The following table lists the components of a multi-clustering definition:

Field	Description
NAME=	A unique character string that identifies the multi-clustering definition. The name field is mandatory and must not match any clustering definition nor search definition names in the same project.
CLUSTERING-ID=	A unique two-character ID prefixed to all cluster numbers. The CLUSTERING-ID field is mandatory. The clustering process verifies the Cluster ID value with cluster-id definition in the CLUSTERING-LIST field.
COMMENT=	Description of the purpose of the multi-clustering definition.
CLUSTERING-LIST=	The string that contains the list of clustering to perform. Separate the clusters with commas and enclose within double quotes. Run all clustering against the same Identity Table. The maximum search value is 16 searches.
IDT-NAME=	A character string that identifies the Identity Table over which the system should perform multi-clustering. IDT-NAME field is mandatory.
OPTIONS=	A list of keywords used to control various search options. Separate each search option with a comma. FULL-SEARCH option specifies that a multi-clustering definition is to process all clusters in the list else it returns on the first cluster that returns some data. When you enable the FULL-SEARCH option, it executes all search defined in the clustering-list in parallel.
SCHEDULE=<list of jobs>	A comma separated list of jobs in queue for multi-clustering. The jobs in the list must be included in the job definition section. The list of jobs should contain jobs related to clustering and post report processing.

The jobs defined in multi-clustering definition should not schedule to perform a LOAD-IDT job.

The following definition lists a sample multi-clustering definition:


MULTI-CLUSTERING-DEFINITION
*======================
NAME=                           multi-clustering
CLUSTERING-ID=                  AA
IDT-NAME=                       IDT-100
CLUSTERING-LIST=                clustering-name,
                                clustering-address,
                                clustering-company
Options=                        FULL-SEARCH
SCHEDULE=                       job-cluster,
                                job-post-all
*
job-definition
*=============
NAME=                           job-cluster
TYPE=                           cluster
CLUSTERING-METHOD=              Merge
*
job-definition
*=============
NAME=                           job-post-all
TYPE=                           post
FILE=                           lf-post-all
OUTPUT-OPTIONS=                 Trim, CR
*

Rename Saved Search

Table of Contents

User Guide

User Guide

Multi-Clustering-Definition

Multi-Clustering-Definition