IBM MDM DATA MODEL BASED ON TYPE(ADDITION, EXTENSION,EVENT MANAGER, External Rules ......)

Data Model
1>ApplicationConfigurationManagement
-APPSOFTWARE,APPDEPLOYMENT,APPINSTANCE,CONFIGELEMENT
2>Custom INQ LVL OR ChildObjectFilter
-V_GROUP,INQLVL,NQLVLGRP
3>Error Handling
-DWLProductType,CodeErrorMessageType,CodeComponentType,CodeErrorReason,Code

SeverityType
4>Event Manager
-PROCESSCONTROL,PROCESSACTION,EVENT,CDEVENTDEFTP,EVENTDEFEXTRULE,EXTRULE
-DAPTERDEF,ENTITYEVENTCAT,CDEVENTCAT,ENTITYEVENTCATOPT,PRODENTITY,
-CDDWLPRODUCTTP
5>Extension Framework
-EXTENSIONSET,EXTSETCONDVAL,CDCONDITIONVALTP,CDCONDITIONTP
-EXTENSIONSET,CDASSERTRULETP,CDDWLPRODUCTTP
6>External Rules
-EXTRULE,EXTRULEIMPLEM,JAVAIMPL,RULEENGINEIMPL
7>External Validation
-V_GROUP,V_GROUP_PARAM,V_GROUP_VAL,V_ELEMENT,V_ELEMENT_PARAM,V_ELEMENT_VAL
-V_FUNCTION,V_PARAM,V_VAL,V_TRANSACTION,EXTRULE,V_ELEMENTATTRIBUTE
8>Notification
-NOTIFICATIONTYPE,NOTIFCHANNEL,JMSCHANNEL
9> Rules of Visibility(ROV)
-CDACCESSORTP,ASSOCIATEDATTRIB,ASSOCIATEDOBJECT,CONSTRAINTPARAM,
-DATAACTION,DATAASSOCIATION,ENTITLEMENT,ENTITLECONSTRAINT,CDOPERANDTP,
-CDOPERATORTP,PARAM_TYPE,V_ELEMENT,V_GROUP,ACCESSORENTITLE,CDACCESSORKEYTP,
-CDPERMISSIONTP,CDDATAACTIONTP,CDFAILACTIONTP,EXTENSIONSET,CDCONSTRAINTTP,
-ACCESSTOKEN,USERACCESSTOKEN,GROUPACCESSTOKEN
10>Search
-SQLSTATEMENT,SEARCHSQL,SEARCHRESULTFIELD,SEARCHCRITERION,CDSRCHFLD
-CDCOMPOPTP,CDELEMENTTP,V_ELEMENT,
11>Security
-USERPROFILE,USERACCESS,CDBUSINESSTXTP
-USERPROFILE,USERGROUPPROFILE,GROUPPROFILE,GROUPACCESS,CDBUSINESSTXTP

12>Historization (sample) Log
-H_CONTACT(All history tables are in the format of the
data tables with an additional 4 attributes:
h_action_code
h_created_by
h_created_dt
h_end_dt)

13>Transaction Audit Information Log (TAIL)
-RANSACTIONLOGERR,TRANSACTIONLOG,INTERNALLOG,
-CDBUSINESSTXTP,CDINTERNALTXNTP,BUSINTERNALTXN,INTERNALTXNKEY,
-INTERNALLOGTXNKEY,V_ELEMENT,V_GROUP,VIEWDRIVER,VIEWINSTANCE,





A configured MDM batch client picks up the files for processing, parses them and invokes the CDI MDM transactions which are mainly custom composite transactions internally calling the MDM OOB txns. All the Party data goes into CONTEQUIV, CONTACT, IDENTIFIER tables, Address data goes into LOCATIONGROUP, ADDRESSGROUP, ADDRESS  tables, Party contact information gets stored in CONTACTMETHODGROUP, CONTACTMETHOD, PHONENUMBER  tables, Account data goes into CONTRACT, CONTRACTCOMPONENT tables, Party and Account association gets stored in CONTRACTROLE and Party to Party associations gets stored in CONTACTREL table, account to account association in CONTRACTREL. Hierarchy data gets stored in HIERARCHY, HIERARCHYNODE, HIERARCHYULTPAR and HIERARYREL tables.

Behavior Extensions

EXTENSIONSET, EXTSETCONDVAL, CDCONDITIONVALTP and CDCONDITIONTP.

Suspect Duplicate Processing (SDP)

1.       CONTACTCDC has all the critical data elements

2.       CDMATCHRELEVTP stores the positive match scores of critical data elements.

3.       CDSUSPECTREASONTP stores the negative scores and

4.       ADDACTIONTYPE table links the various match and non-match score permutation to a suspect category

5.       SUSPECT to store the score

Tree-style hierarchies

A Tree-Style hierarchy structure can be defined between the elements of a set

HIERARCHY, HIERARCHYULTPAR, HIERARCHYREL, HIERARCHYNODE

Mapping Between Two Data Sets

Association between two data sets can be done by using Mapping. Each mapping has its own metadata, including version and lifecycle state. For instance, the version for a mapping is unrelated to the version of the sets that are being mapped. Mapping can act as a cross reference between source and target sets. Mapping Data Type can be created same as Reference Data Type.

RDVALUE, RDVALUESETRELATIONSHIP, RDVALUERELATIONSHIP, RDVALUERELATIONSHIPPROPERTY

CREATING REFERENCE DATA TYPES Every data set and mapping is associated with a data type definition. Only users having Administrator role can create a reference data type. Tables affected are RDVALUETYPE and RDVALUEPROPERTYTYPE

LIFECYCLE AND STATES A set of lifecycle states is defined for a reference data set and mappings, and transition rules determine what the valid state transitions are. The set of lifecycle states and allowable transitions from one state to another can be defined and changed to support different review and approval process flows. Some of the common life cycles are:

Simple Approval Process (One approver needed)

Two State Approval Process (more rigorous governance controls, Multistep process)

Active Editable (No Approval process, state property disabled)

State Machine 2 (only 2 states, No need of separate approver, same can do)

RDSTATEMACHINE defines the lifecycle for a set or mapping with its initial and final state. Transition from one state to other in a life cycle is defined in RDSTATETRANSITION with allowed action. Such As Draft->Pending Approval ->Approved/Reject ->Dropped. Each Value Set will have one lifecycle. After the set is approved data cannot be imported into the data set. Any custom state should be entered at CDRDSTATETP and action at CDRDACTIONTP.

CREATING REFERENCE DATA SET Each reference data set is based on a reference data type. The reference data sets contain your reference data. You can create new sets, edit existing sets, delete unwanted sets, and manage versions of the reference data sets and set hierarchies. You also provide the reference values for the sets and translation values for them.

Tables involved are RDSET and RDVALUE


http://www-01.ibm.com/support/knowledgecenter/SSWSR9_10.1.0/com.ibm.mdmhs.comn.dict.doc/r_SubjectAreas.html