DNB - Den Norske Bank (KVH 2.

0
Application)

Application Understanding Document for Konsern VareHuset

KVH 2.0 (Enterprise Data Management)
Project: KVH 2.0 AUD for KVH 2.0

Document Control
Revision History

Date Version Author Description

Distribution

Name Title Document Version Date

Approval Signatures

Name Title Document Signature Date

Version

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 2 of 126
Project: KVH 2.0 AUD for KVH 2.0

Table of Contents
1. BUSINESS.......................................................................................................................................................5
1.1. BUSINESS OVERVIEW......................................................................................................................................5
1.2. BUSINESS PROCESSES.....................................................................................................................................5
2. FUNCTIONAL OVERVIEW........................................................................................................................6
2.1. ARCHITECTURAL AND FUNCTIONAL OVERVIEW................................................................................................6
3. TECHNOLOGY OVERVIEW & COMPONENTS..................................................................................7
3.1. TECHNICAL OVERVIEW AND COMPONENTS OF KVH 2.0 :.............................................................................7
3.2. ENVIRONMENTS..............................................................................................................................................12
3.3. DIHUB WALKTHROUGH..................................................................................................................................29
3.4. DIHUB INTERFACES (PRIME)...........................................................................................................................37
3.5. DIHUB INTERFACES (HR)...............................................................................................................................37
4. LOGICAL AND PHYSICAL MODEL...................................................................................................................37

4.1. ERWIN DATA MODEL.....................................................................................................................................37

4.2. DATA MODELS IN KVH 2.0.............................................................................................................................41
5. REFERENCES AND DOCUMENTS...................................................................................................................47
5.1. REFERENCES..................................................................................................................................................47
5.2. DOCUMENTS..................................................................................................................................................51

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 3 of 126
Project: KVH 2.0 AUD for KVH 2.0

OBJECTIVES

The purpose of this document is to record all the knowledge and information that has been
captured during the knowledge transfer sessions and job shadow/reverse job shadow phases of
the knowledge transition, held between 18 th March 2025 to 18th April 2025. This includes all the
detail and topics as per the “knowledge objects” column in the associated Knowledge Acquisition
Plan (KAP).

This document is to be used as a reference point for the Capgemini knowledge recipient once the
sessions have been completed and as such doesn’t replace application/system documentation or
any existing application handbooks/manuals. For completeness, this document has been
reviewed by the respective SME for accuracy and correctness. “

 For preparing, please examine all already existing information’s from application data list,
transfer them into the document and check them together with Cluster Lead of DNB
(KVH 2.0)

 The KT approach is defined and documented in the <share point folder name>

 If information is avalable in existing documentation, please insert a fully qualified Link to

the document. It is not necessary to document existing things again

 The templates will be maintained and optimized during the KT for AMS, so please use
always the newest template stored in the sharepoint

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 4 of 126
Project: KVH 2.0 AUD for KVH 2.0

1. BUSINESS

1.1.BUSINESS OVERVIEW
KVH 2.0 is an Enterprise Data Warehouse (EDW), where business
services and business users are under one or more business
capabilities, where the services are orderable by users, internal or
external customers via Service Catalog.
Description: Provisioning the corporate data warehouse as a
trustworthy source of information for the DNB Group.
Mission: Run and maintain the enterprise data warehouses and
support key processes in DNB.
Stakeholders:
 Business Owner: PDI (Product Data and Innovation) Data
division (Isabel Barroso-Gomez)
 IT Owner: T&S EDW EDM tech. family (Kathrin Volkmer)
 Users: DNB group wide
1.2.BUSINESS PROCESSES
 The Business offerings consist of platforms like IBM
Information Server (ETL Platform), Teradata platform, Core
Infra (Windows, Linux, DB2, MQ), Cloud (Azure, AWS) are
also used.
 Applications include KVH2 batch, DI Hub, Basel Calc batch,
RDM, DM6.
 KVH2.0 Datawarehouse, selected Teradata apartments,
and applications like profitability reporting, international tax
reporting, audit reports, etc. are the service offerings.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 5 of 126
Project: KVH 2.0 AUD for KVH 2.0

2. Functional Overview

2.1.ARCHITECTURAL AND FUNCTIONAL OVERVIEW

Architecture Overview of KVH 2.0

 The Data Sources (left part of above figure) consists of different tables like
Transaction, Organization, Scoring, Event, Case, etc.
 The types of data are mostly retail, wealth and corporate banking data.
 The other components which are necessary for KVH2.0 are ETL, DI Hub,
RDM, Erwin, and Teradata.
 DI Hub and RDM together Process and Load into 1300 KVH 2.0 Teradata
Tables.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 6 of 126
Project: KVH 2.0 AUD for KVH 2.0

3. Technology Overview & Components

3.1. TECHNICAL OVERVIEW AND COMPONENTS OF KVH 2.0 :

DI Hub (Datastage)
Teradata
RDM
i-Know Tool to be used for scheduling in Lower Environment
Autosys Tool to be used for Scheduling in Production

a) DI Hub: BA01219 DI Hub

 Data Integration (DI) Hub is the unit in which all the jobs, principles for how
the jobs are designed and executed.
 Party data is coming from IAP (Inbound Access Point) to inflow (DS jobs are
present) and to OAP (Outbound Access Point).

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 7 of 126
Project: KVH 2.0 AUD for KVH 2.0

 DI Hub is an event – driven architecture, independent of the batch

framework, and stateless. (no dependencies on other jobs)
 Components of DI Hub are IAP, Dispatcher, HIR (Table Definition), HCR
(Hub Canonical Record), Inflow, OAP.

Design Principles of DI Hub:

b) Teradata : Business Application BA01300 KVH2.0 Overview

 The Datawarehouse of KVH2.0 consists of 25% of Teradata capacity.
 The main components of KVH2.0 are the physical warehouse
implementation of Teradata (tables, views, procedures and related
metadata), Automation Studio, Smart Admin, and Authorization custom
application.
 Apartments is a term in Teradata where it consists of lot of databases.
 KVH2 was built in 2010.
 We use Teradata Temporal feature for loading the data, where each table consist of
complete history of the data.
 The Teradata is continuously loaded in KVH2, parcels are loaded as soon as they are
available rather than loading daily/weekly/monthly/ batches. (Higher frequency/smaller
packages).

Elements of typical KVH2.0 Teradata SQL Pipeline are:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 8 of 126
Project: KVH 2.0 AUD for KVH 2.0

 Party data (Customer Data) are equal in KVH2.0 and KVH1.0, agreement
data are the different sources of KVH2.0 and KVH1.0.
 The data is sourced from DI Hub to Load table for load procedure.
 Then the control is handed over from DI Hub event model to KHV2.O
Smart Admin event model.
 Then the Smart Admin runs the Flush Procedure which updates the target
table.
 Then we have multiple views in KVH2.0.

Design Principles of KVH 2.0:

 Authorization tools authorize all the tables, those are running on Teradata.
 Every database object which exposes data must be classified with a data
authorization category.
 Every database object must be categorized with subject area and row level
access control setting.
 Every column must be classified with a column category for masking.
 All Data has relationship to time (has the timestamp column, which is
automatically added by the metadata tool i.e., Automation Studio).
 All data shall have a defined lifecycle policy based on the timeline.
 Referential Integrity are the rules which ensures consistency and
validation of data. (-1 = Unknown or Empty, -2 = Invalid, -3 =
Unknown non-empty by approved).
 Physical data model shall be optimized.
 Everything in KVH2.0 is always created in SDM (Erwin Model), we update
the model first then update metadata in Automation Studio.
 Loading of the EDW is an automated process.
 Manipulation of data in the production database is not allowed.
 EDW is built on three schema concept (Physical Layer [tables], Conceptual
level [Simple view], External level [Application view]).
 KVH2.0 uses Natural Key.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 9 of 126
Project: KVH 2.0 AUD for KVH 2.0

Simplified View structure in Teradata KVH:

Where we have no access to the table, and there is a Trivial view created for tables, which is
called as the P_Wt (one to one views). And the top layer consists of views created with added
layer of security for the middle layer. (P_WtSec). The Tables are sorted based on the
environments, example – Production (P) or EDW.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 10 of 126
Project: KVH 2.0 AUD for KVH 2.0

KVH2.0 Overview:

 Authorization: It is used for checking all the necessary access for the data.
 Smart Admin: It is an event driven orchestrating procedures and
monitoring.
 Automation Studio: Based on the information model (SDM), produces DDLs
to be deployed in Teradata.
 Teradata platform: for data sourcing and writing.

c) RDM:
 Reference Data Management Hub (RDM) is used for reference data
maintenance which can be later accessed as for lookup.
 It is mainly consisting of all the required master data for lookup. (to be
decommissioned soon)
Design features of RDM:
 It is known for its governance and stewardship of hundreds of master sets,
and thousands of source sets and mappings.
 Providing metadata maintenance, versioning, policy management and
workflow support.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 11 of 126
Project: KVH 2.0 AUD for KVH 2.0

 It is a metadata driven event-based synchronization to KVH2.0. (even for

external reference data sources)

RDM current architecture:

 The current architecture consists of IBM RDM where the reference data is
sourced from the database DB2.
 The DI Hub processes the ETL Jobs where the master transaction data and
the lookup data is compared, with populating Redis in the background
using operator (C++)

RDM target architecture (2025):

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 12 of 126
Project: KVH 2.0 AUD for KVH 2.0

 The majority of the RDM content will be migrated to Cinchy.

 Supporting by redeploying all the target objects on Teradata, lookups, and
testing.
 Will be using DB2 for the target architecture too, by use of complex data
pipelines, jobs, etc.

d) I-Know Tool:
It is a scheduling and monitoring tool used to monitor ETL Runs/Process
Managements/To add dependencies on upstreams and downstreams/Logs
e) Automation Studio:
We have two types of Automation Studio access, one is onsite VDI and another
one is Windows 11 (like a package) under the VDI.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 13 of 126
Project: KVH 2.0 AUD for KVH 2.0

The application on the left side here is talking to web service and we have the Windows
server in the middle. We have two servers which handles different environments.

KVH2.0 as seen from Automation Studio:

The load table, the procedure, the flush procedure that moves data over from load into the target
and all the views that build shows data here to the end user and the applications. Everything
here is created by Automation Studio by using templates.

ASL Code Templates:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 14 of 126
Project: KVH 2.0 AUD for KVH 2.0

Table Types of Automation Studio:

Set of features or Templates of specific tables in AS:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 15 of 126
Project: KVH 2.0 AUD for KVH 2.0

How to Start AS:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 16 of 126
Project: KVH 2.0 AUD for KVH 2.0

Automatically created objects:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 17 of 126
Project: KVH 2.0 AUD for KVH 2.0

3.2. ENVIRONMENTS

We have 4 environments in KVH 2.0 both for DataStage and

Teradata
DEV
STest (SIT Testing)
ATest (UAT Testing)

PROD
The Confluence URL to See the Server details for Dev/Test/Prod are
given below:
Server Mappings - P-9159: Datastage ETL Upgrade - DNB
Confluence

DST (Dev Environment) :

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 18 of 126
Project: KVH 2.0 AUD for KVH 2.0

SIT (Test) :
Same server but two projects one for SIT and other for UAT
Note:
vip 1: used for kvh1
vip2: used for kvh2

Prod:
Prod1: kvh1
Prod2: kvh2

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 19 of 126
Project: KVH 2.0 AUD for KVH 2.0

(a) Datastage:
Datastage Folder Structure according to DI Hub Job Flow:

Inbound Inbound Outbound

Access Point Flows Access Point
(IAP) (Inflows) (OAP)

IAP
Initial jobs are like Staging jobs where initial data will flow/load and
create datasets. Datastage jobs will read Source which could be
Input Files/Third party database (tables) or within Teradata itself
and load it to staging area (datasets) which will act like source for
InFlows DS Jobs.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 20 of 126
Project: KVH 2.0 AUD for KVH 2.0

INFLOWS:
Some of major transformation will happen here through datastage
as shown below. It will also generate datasets which will be source
for OAP DS jobs.
Note: Initialize current files will be used for Delta logic.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 21 of 126
Project: KVH 2.0 AUD for KVH 2.0

OUTBOUND ACCESS POINT:

Mostly we will load data into KVH 2 (Teradata) which are called
Load Tables which will load data through datasets generated in
inflow jobs and then through flush procedures it will load data into
target tables.

In the same job, inside “After SQL” properties of Teradata

connector, we will call some procedures called as Flush Procedures
as shown below:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 22 of 126
Project: KVH 2.0 AUD for KVH 2.0

(b) iKnow: (Dev/SIT/UAT) : Monitoring/Scheduling Tool

The GUI looks like below:

Director: It acts like a monitoring area

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 23 of 126
Project: KVH 2.0 AUD for KVH 2.0

Event Management: Some jobs are event driven (File

driven/MQ/Smart Admin schedule)

Process Management: Acts like Scheduling Tool in Lower

Environment.

There are two Environment for test Server as shown below:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 24 of 126
Project: KVH 2.0 AUD for KVH 2.0

S-Test
A-Test

(c) Teradata (1300)

It has only two server which is test server which have

First Server : Dev (D_)/STest (T_)/ATest(Q_)
Second Server: Prod (R – Prod Like (A_) and P-Prod Env (R_))
Note: A_ is obsolete now

Naming Convention of Schemas in Teradata:

D_EdwStg : Staging Layer data
D_EdwLoad: Load( Stg like)
D_Edw : Target Table/Main Table ( Being loaded through Flush Proc from
EdwLoad)
D_EdwTmp: Temporary Database
D_EdwMd : Manual entries
D_EdwOwner: All procedures are stored in this Schema

Load procedure : _LP

Flush procedure: _P

f) RDM: dev/test/prod

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 25 of 126
Project: KVH 2.0 AUD for KVH 2.0

Prod Login URL: RDM : https://prodd.dnb.no/RefDataClient/public/LoginPage.jsp

Sit Login URL: https://testa.dnb.no/RefDataClient

Source Sets: It contains Key and Mapping sets (with Master sets)

MAP_Curency_0238 : It will start with MAP then RDM Set name (_Currency
here ) and Source (_0238 here)

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 26 of 126
Project: KVH 2.0 AUD for KVH 2.0

g) Linux Server
Common Path : /dnb/prosess/InformationServer/
ATest: UAT Env
STest: SIT Env
Dev: Dev Env
Prod: Prod End

Folder Path inside ATest:

/dnb/prosess/InformationServer/ATest

NR1219: This is the path where all incoming and outgoing files will be available

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 27 of 126
Project: KVH 2.0 AUD for KVH 2.0

in: All incoming files

ut: all files needed for NFT team to process
out: All outgoing files

Above figure shows the path where all the incoming files will be landing
This is the path where file dispatcher runs to process the file and when event
handler runs it will trigger DS Jobs.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 28 of 126
Project: KVH 2.0 AUD for KVH 2.0

MFT team which is a third party system within DNB, they will fetch file from above
path and deliver it to target team. After successful deliver they will put copy of the
files in data arch (mentioned below) path.

DIHub folder in Unix:

All the job configuration and parameter needed for DIHub to run is present here in
the below folder structure. Let us understand use of

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 29 of 126
Project: KVH 2.0 AUD for KVH 2.0

Scratch: Use this path for Scratch disk space

Temp: We are not suppose to use this folder
Current: To hold the current DS for SCD Type change for the code
Datasets: Default Datasets path used in .apt at the project level
Input: Input file for our references
Output: Output file for our references
CustomComponents: To define shell script reauired for particular design resides
here, 1300 monitoring script resides here etc.
AutosysDispatcher: it has script through which DS jobs triggeres pearl script
FileDispatcher: it is very important . Event handler creates a parcel directory.
For every execution we have a dedicated parcel directory created by event
handler as shown below :

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 30 of 126
Project: KVH 2.0 AUD for KVH 2.0

So every execution of Datastage jobs will get its own dedicated parcel directory.
And all the HIR and Source file will be created under above dedicated parcel
directory.

Configuration File: This is the most important file in an event based system.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 31 of 126
Project: KVH 2.0 AUD for KVH 2.0

Configuration files in the folder shown below:

Configuration file for file arrival event:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 32 of 126
Project: KVH 2.0 AUD for KVH 2.0

Configuration file for KVH Job: it is not fired through File arrival event.

g. Automation Studio:
 It is a automatic deployment tool for deployment of DDLs, views,
Stored procedures in respective environments. Works in prod
server or devtest lab and decommissioned for Dev.
 For Code generation, 95% of KVH2 is based on templates, same
structure for every table in the data warehouse.
Note: if you face any error while connecting to it, connect with DAB team
on this.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 33 of 126
Project: KVH 2.0 AUD for KVH 2.0

Open EAT: Used for deployment of any new table. Provide Erwin model
into this and this will do automatic deployments. It will automatically give
unique name to tables, Column name and index name.
Configure tables: Provide different authentication at the tables level,
column level etc. as shown below:

For Deployment to Higher env, click on Export to XML option as

shown below:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 34 of 126
Project: KVH 2.0 AUD for KVH 2.0

Flow Diagram of Models :

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 35 of 126
Project: KVH 2.0 AUD for KVH 2.0

(h). Smart Admin: Status Monitoring tool for Teradata (1300)

Scheduling component running constantly, it receives events and kicks off the jobs
and can also kick off jobs on a timely manner, like a crude job schedule.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 36 of 126
Project: KVH 2.0 AUD for KVH 2.0

Detailed explanation how the Smart Admin works:

The DI Hub sends the data to the Queue Table, which are in MDS extension database and
then, the Smart Admin will create the instantiate jobs where it collects the statistics on
the load table, then will run the flush, does the cleanup so the flush done event is
completed.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 37 of 126
Project: KVH 2.0 AUD for KVH 2.0

Example of Job Dependencies on Smart Admin:

Where the specific actions are carried out and triggered, jobs waiting for other jobs to
complete, etc.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 38 of 126
Project: KVH 2.0 AUD for KVH 2.0

Scheduling Jobs in Smart Admin:

Smart Admin features:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 39 of 126
Project: KVH 2.0 AUD for KVH 2.0

One Special feature of Smart Admin is that the job group is having the jobs that are
running serially, but two or more different job groups are running parallelly.
Note – The Main job group queue can run 5 jobs serially.

i). KVH2 Auth:

It is the authorization tool which reads the data from the active directory
and translates it into access to roles on Teradata. (It is difficult to
understand)

j). Auth Info Self Service:

It is a web-based lookup for KVH2Auth, which is easier for understanding.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 40 of 126
Project: KVH 2.0 AUD for KVH 2.0

Link to access Auth Info:

https://authinfo.aac.dnb.no/

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 41 of 126
Project: KVH 2.0 AUD for KVH 2.0

k). VedApp:
Mostly used by KVH1, but sometimes we use this in KVH2 for copying the
small amount of data between environments.

l). Teradata Viewpoint:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 42 of 126
Project: KVH 2.0 AUD for KVH 2.0

3.3. DIHUB WALKTHROUGH

Approach of Event Driven Datawarehousing and its architecture

DIHub is purely event driven architecture. It means everything that happens ,
happens because of event.

Architecture

 Dispatcher actually fires the event through Pearl Script and

then these events triggers the DS jobs (Not Mandatory
always). These dispatchers are responsible for setting up
the data provider code, setting up the business date and
other housekeeping columns.
 Dispatcher are the action items that creates a data packet
and then create an event.

There are multiple types of dispatcher:

Prime Dispatcher/ODS Finance dispatcher (Source as ODS)
MQ Dispatcher: Checking for Messages in MQ
KVH Dispatcher (EDW Dispatcher): Source is KVH
File Dispatcher: Sources are file

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 43 of 126
Project: KVH 2.0 AUD for KVH 2.0

There are multiple types of Events:

File Arrival Event (FE) : Triggered when file arrives (By MFT
Team)
MQ Message Event (MQ) : Triggered when Message arrives
(Every 2 hrs. looks for messages). E.g. Customer data in
KVH 2
General Event (GE) : Event without a data packet to define
a dependency or to delay some process to run.
Composed Event (CO) : An event that can use multiple
events. E.g. DS job required 2 files from a source so in this
case we will define composed events for 2 files to arrive.
Once all the events that are required for composed events
are processed then next job will be processed. It can be
defined or handled in configuration file.
All above functionality is driven through a tool called i-Know
Tool.

Life Cycle of Event:

An Event is Fired Event Handler will Run Event
handler(In iKnow) will see all the events which have been
fired  It will pick the fired eventit will look in the
configuration file what datastage job to trigger for that
event which is firedprocess the eventcheck for event
which have been processed UI is updated.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 44 of 126
Project: KVH 2.0 AUD for KVH 2.0

Parcel Directory : File Dispacther generates the event for

that particular Parcel_ID and that Parcel_id will be unique
for each execution of the process.
MFT : putting the file at the correct path
File Dispacther: it is picking the file and firing an event
Event Handler: it is processing the event
And then datastage job is executed.

Two important thing we need to be careful about is:

1. How we design datastage jobs
2. How we define configuration file.
(2) DS Folder structure :

All folders are created under the area that we will be working
with.
In the above figure, we have created folder under DataArea, then
Inside InboundAccessPoints we have folder based on different sources.

We have a folder with the name InboundFlows for Main DS jobs

where we will be applying transformations and business rules.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 45 of 126
Project: KVH 2.0 AUD for KVH 2.0

We have a folder with the name OutboundAccessPoints where all

the outbound DS jobs will be present. It may have source based jobs
where we want to export files to downstream otherwise other jobs will
not be source specific. If our target is Teradata , then you will found
those jobs in KVH2.0 subfolder here.

Dispatcher specific jobs for ODS dispacther, MQ Dispatcher and EDW

dispatcher, we are the one who takes care of it that means we are
responsible for collecting data and creating an event out of it. And the
folder for same is
DiHubGeneralJobs  Autosys_Dispatchers as shown below:

ODS Dispatcher: We have two kinds of ODS dispacther

1. Finance Dispacther
2. Prime Dispathcher.
Finance and Prime are two different sources.

Execution order of DS job in DIHub:

IAP_SEQUENCER (DS parallel Jobs for IAP and inflow both) -DS
Job Outbound
IAP_SEQUENCER is executed through Config files.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 46 of 126
Project: KVH 2.0 AUD for KVH 2.0

Default.apt: it tells DS what resource disk and what scratch disk a

DS job will run and on how many nodes a DS job will run.

Naming convention to be noted:

IAP Jobs have sequencer and it ends with _S as suffix.
OAP jobs doesn’t have sequencer and it ends with _J as suffix.

IN KVH 2 , We have two ODS dispacther :

1). PRIME (dispatcher checks every 15 mins)
2) ODS FINANCE

IAP SEQUENCER:
All the jobs in IAP sequencer and OAP jobs must be multi sequencer
jobs to be handled by event handler.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 47 of 126
Project: KVH 2.0 AUD for KVH 2.0

Parameters for IAP Sequencer Job :

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 48 of 126
Project: KVH 2.0 AUD for KVH 2.0

To Create an event we need three parameters to be used (reference values)

ReferenceValue1: batch number
Reference Value 2: Some housekeeping column
ReferenceValue3: FULL or DELTA
Parcel Id as a parameter will be allocated dynamically and every execution will
have a unique Parcelid.
DataExtractCd : to run the job for DELTA (D), History or FULL(F)

For Prime : It will always be DELTA

For CDC logic to work, We have to set DataExtractCd as F

IAP Job:
Details regarding Mapping and all, we will get that in Inbound Mapping
specification sheet

SOURCE TRANFORMER HIR

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 49 of 126
Project: KVH 2.0 AUD for KVH 2.0

Standarisation to be used in AIP Jobs:

svRecordId : Unique record number and below housekeeping columns:

Below are two most important housekeeping columns:

LastChangedDttm
LastChangedDttmTZ
DLastBusinessDttm
dnb.TrimNonPrint : For String column always use this.
Dnb.StringToUTCTimestamp

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 50 of 126
Project: KVH 2.0 AUD for KVH 2.0

Shared Container (Above Picture of IAP DS Job) : It loads data to staging

table (P_EdwStg) in test Env as in production use of this is prohibited (shown
below) :

Job Inside Shared Container:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 51 of 126
Project: KVH 2.0 AUD for KVH 2.0

DNB has defines some custom routines:

For Every Time Zone issue, Use timezoneOffsetSeconds = dnb.use Timezone

(“TZ”)
Note: We are using DNB specific Routines to keep standardization across KVH 2
Example:
Irrespective of any timestamp format from the source, is we will use DNB Routines
and call that in our DS jobs it will handle such different timestamp format coming
from Sources to convert into same timestamps:

Example:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 52 of 126
Project: KVH 2.0 AUD for KVH 2.0

While Running the IAP Sequencer, we will run the PARTY HIR to HCR first and then
rest of the HIR to HCR to avoid FELIX call to happen multiple request in same time
for the same customer . In other words first we will be process customer, then we
will get all the details in CACHE for the new customer and then rest of everything
can be executed subsequently.

In the above snippet, We have kept dependency ion event in such a way that
Agreement job will be executed once Party Area is processed.

INFLOW JOBS:
Developed by Iknow Utilities having custome built stages written by C++.
There are multiple custom built stages developed in DNB :
1. RDM
2. GKM Lookup
3. Felix Stage

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 53 of 126
Project: KVH 2.0 AUD for KVH 2.0

4. Fine Delta Logic

OAP JOBS: It takes Input as HCR DS.

Every OAP job in KVH2 is a bulk load. That means in Bulk load, we drop all the
indexes in the WORK Table and delete data of that data in PARCEL_ID already
present. This will ensure that our OAP jobs are restartable in case of partial load in
previous runs (shown below):

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 54 of 126
Project: KVH 2.0 AUD for KVH 2.0

In After SQL, we are calling FLUSH Procedure to load Target table from Load Table.
3.4. DIHUB INTERFACES (PRIME)
 Some of the observations and exports will deliver based on prime.
 DI Hub there are two kinds of export. One is where we are kind of a mediator between
downstream application, and we work as data provider to downstream application.
 And we provide purely on source data to downstream and we do not intend to change any
data.
 One export which we give to
 Transaction team which is purely based on the data we receive from Prime.
 exports are derived directly through the IAP stage and then we trigger an event which
triggers export job and deliver to downstream, application.
 if we take an example of efraud which is directly give the flex data to downstream
application.
 It means that we do not use KVH data to deliver majority of our export.
 We deliver data from KVH to data and then we deliver that data to downstream application
and sometimes we share views that are built on top of KH2 table which we use and
eventually deliver data to downstream application.
 If u see prime doesn't have any system to deliver data to IPA transaction lake.
 And Transaction is a financial event that’s why we called it as financial event because we
can’t hear any word transaction in kvh2.0 table.
 If we see any financial event, we can understand that is a transaction event.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 55 of 126
Project: KVH 2.0 AUD for KVH 2.0

If we take an example for Money transaction:

Money Transaction is a IAP sequence for the transaction job event, which we are triggering this
event for this export so it is simple process there is no inflow or any other DI hub specific
transformation anything for that ,just simply reading data from source and create and event and
that event deliver the data, but we are using DiHub principles for extracting data from
source .Money Transaction is a IAP sequence for the transaction job event, which we are
triggering this event for this export so it is simple process there is no inflow or any other DI hub
specific transformation anything for that ,just simply reading data from source and create and
event and that event deliver the data, but we are using DiHub principles for extracting data from
source .

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 56 of 126
Project: KVH 2.0 AUD for KVH 2.0

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 57 of 126
Project: KVH 2.0 AUD for KVH 2.0

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 58 of 126
Project: KVH 2.0 AUD for KVH 2.0

If we see below screen shot, created data set directly and reading data form source and then
extracting data by triggering an event after created an event.

If we see HIR implementation, that columns, we are receiving from prime and sending to
downstream like target and we are not responsible for any data related issue.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 59 of 126
Project: KVH 2.0 AUD for KVH 2.0

If we see above ss design flow, we created a dataset in the IAP job and then create the event,
that event it is the DIB host data set job we trigger that executed eventually and this is file arrival
event it means it files to be transferred.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 60 of 126
Project: KVH 2.0 AUD for KVH 2.0

If we observe above screen shot there will be job name:DIHub post dataset and invocation id :is
prime Transaction event and under parameters : schema name: Hir_Transaction_event and data
provider code and business date etc.. so we are passing to downstream event by OPA job .

 Incoming files we cannot say directly from source same way out going files also not directly
send to Target there is intermediator between those like MFTS systems ,down steam to ritm
and ritm to MFTS and Autosys also involved if it delivery within DNB then It Taken care
Autosys or if it is from external MFTS get involved ,So MFTS put it in specific path that
standardized and deliver to downstream application.

Win SCP path:

WinSCP is used for placing files in standardized path like UT.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 61 of 126
Project: KVH 2.0 AUD for KVH 2.0

If take one job like below:

That config file has specific path and placed, and it contains some job and schema name path,
once it triggered, we can see from where it got triggered based HIR Hub dataset.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 62 of 126
Project: KVH 2.0 AUD for KVH 2.0

CMM file formatter:

Some of the job will maintain the Formatter in yellow highlight below ss. which build based on
Java code so which triggers basically data stage stages to java code and it contains some Custom
properties like Source system name and parameters mentioned.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 63 of 126
Project: KVH 2.0 AUD for KVH 2.0

Refer below confluence for more info about CMM file formatter

CMMFile Specification - 1219 DI Hub - DNB Confluence

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 64 of 126
Project: KVH 2.0 AUD for KVH 2.0

Export job:
Export ,If we takes CMM export prefer two kinds like If source have specific requirements then we
can create the custom report and other one generated for kvh2 its sometime scheduled on
based time and that particular time smart admin event will be fired and our EDW dispatcher will
read that event and trigger the export job else through smart admin schedule an flushed event
base.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 65 of 126
Project: KVH 2.0 AUD for KVH 2.0

Directly pulling from KVH table /view below ex ss:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 66 of 126
Project: KVH 2.0 AUD for KVH 2.0

3.5. DI HUB INTERFACES (HR)

 DIHUB HR contains sensitive data that is related to associate or employee related things
for which.
 We have like migrated all the jobs related to associate and employee to this DI of HR
where we have like access restricted.
 Only KVH 2 and platform got the access restricted by using a general group with help of
EU support team where all our AB IDs will be under this group.
 And whoever the joins in that specific to kv2.0 they get access along with DI HUB and our
environment so this Restricted access can be contains 3 applications, such has data stage
& Linux and AutoSys.

If we see above ss for I know there will be 2 tabs

1) DIHUB
2) DIHUB_HR

We have paths similarly DIHUB_HR like DIHUB ,only difference is prod HR instead of prod like
below ss.
So different paths:
Config file: All the related DIHUB HR related files will be resides
Source landing: Where all the source files related to HR files landed in this path.
And Target _archive and Target and object restore and shell script path.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 67 of 126
Project: KVH 2.0 AUD for KVH 2.0

If we open the Autosys, there we can see DIHUB AND DIHUB HR together under views Related
jobs and their status.

ViewsDIHUB-1219DIHUB AND DIGUB related DISPATCHERS we can see below so first half
would be DIIHUB AND second half would be a DIHUB HR.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 68 of 126
Project: KVH 2.0 AUD for KVH 2.0

And for login to DataStage with DIHUB HR sever like below.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 69 of 126
Project: KVH 2.0 AUD for KVH 2.0

If we see in DIHUB HR there will be four categories mainly observed below ss like Data area

Data areas:

Data source sources that is related to this associate data that is one is like answer data contact
center and simply HR and team jobs.

Payroll:

It is associate or employee data that we have inbound access point jobs like IAP jobs

Contact center:
Contact center has around 650 plus jobs that we have migrated from DI hub to DIHUB HR
And the others are like 10 jobs, and we have like majority of the files would be like CMM files
related and the export files that we deliver to the downstream.

The Job developed them same as how we do it in DI Hub.

For IAP, we use the same template and for inbound and initialized job we use the Ino and for the
OAP outbound job it's the same.
everything is same, but only the landing and the Target path is different, and the access is
restricted.

If we required more info and we refer below confluence:

DIHUB HR - 1219 DI Hub - DNB Confluence

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 70 of 126
Project: KVH 2.0 AUD for KVH 2.0

3.6. DB2 AND REDIS

 DB2 is not only important for process any sources, but also for processing all the
metadata and the utility related data.
 There are several custom procedures and views are created at the DB2 level. (example –
people related data)
 MDM DB1 will no longer be required for us as we are about to move out of RDM by 2025.
 The data which are needed for custom development apart from regular RDM data
residing in MDM DB1, will be migrated to IM DB1. (Currently using both)

Tables in DB2:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 71 of 126
Project: KVH 2.0 AUD for KVH 2.0

 Dispatcher.event_status, diaspatcher.job_status, dispatcher.parcel are the table names,

present in the DB2. (Dispatcher_hr.event_status for HR)
 We have data in IM DB1 for GKM (Generic Key Manager) as well, it is the key manager for
creating, and map unique keys, generated within the system.
 GKM component basically consists of C++ code that looks back-end data in DB2, which
requires parameters like Agreement system, key agreement, reference key, which are
defined by data modeler in inbound mapping specifications.
 The IM lookup and the GKM lookup primary check is done in the Redis cache.
 The Keys are responsible for data retrieval from any domain or source and should be
carefully referenced while mapping the data. The parameters should be taken care to be
given properly.
 Domain is nothing but the type of data is sourced from. (example Agreement, Asset,
Promotion, Quotation, etc)
 ID will have the domain attached to it whenever it is generated.
 Metadata for all the dispatcher is also present in this database.

ODS Data Source:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 72 of 126
Project: KVH 2.0 AUD for KVH 2.0

Finance ODS 1602 data warehouse. This is the Finance data warehouse. Maintained by DXC
team.
Mercur (4010) and Origo (4027) are the data stores where the DNB gets the data from.

Servers of ODS databases (read-only):

We are just providing the data from the sources to the RCM team for reporting.
Note – We use ODBC connection for this database, in SQL Assistant.

Source sets and Master sets for Mercur and Origo:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 73 of 126
Project: KVH 2.0 AUD for KVH 2.0

ODS Finance Confluence Link:

https://dnb-asa.atlassian.net/wiki/spaces/DIH/pages/152404104/
DI+Hub+Source+1602+Finance+ODS

3.7. RDM (TO BE COMPLETED BY VENKAT)

RDM:

IBM RDM application is a web-based application that we use and also this application is maintained by platform
team and which is the 110 routine of about DNV.

As part of the development activity, EDM and kvh2 team is manage this code and all the environments.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 74 of 126
Project: KVH 2.0 AUD for KVH 2.0

For more information we can refer below confluence:

4410 Reference Data Management - DNB Confluence

whenever we start a project and if you have any requirement of adding RDM lookups on that.
Example:
If we are working on any source data and source file which has any country codes and other codes, so our data
modeler or architect will create the IAP mapping specification and

We have tab in RDM, from there they will fill up information and they order the reference codes.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 75 of 126
Project: KVH 2.0 AUD for KVH 2.0

If we take an example for codevork , we get the business request from the jan-egil for RDM and EDM team
and other tech families as well like AML, FEC, RCM, so data molder will take action on their request and let
them completed as their standards follows, it should go through this code ever mail chain they update .
So They order or update the master set request on SharePoint and then they send us mail on code event and
then from there we take it forward.

So marinating this maser sets comes under codever responsibility in all environments like prod and test.
 In test environment we have source sets and mapping usually taken care by project team in lower
environment, but prod, plat form team will be taken care.
So when you raise the crq you add a task for implementation for platform team and we should guide them
what they needs to be do, similarly they should follow the same step for prod as well.
master sets will go through codeverk only.
Below share point link is use for reference, how we can order the master set code and reference set name
Test RDM Link - IBM Reference Data Management Hub

RDM - New reference data orders .

And as a developer no need to do anything but as kodevrk team we should guide them the steps.

If we take an example for latest master data set ,it contains like Duration type and content type and order type
like New master set and their reference set name and duration type and project id and then who ordered that
person name and different process statuses under that we can choose like prod ,Atest ..etc.. They will update.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 76 of 126
Project: KVH 2.0 AUD for KVH 2.0

And Each and every master set which is modelled in RDM is also modelled in KVH 2 using the same process
we follow for our continuous or any other tables. We followed the same process for RDM as well go through
SIM, SDM PDM and EAT and Automation Studio.

So they will update if the SIM modelling is done or not and then they can update the e-mail and these are like
not mandatory things but if we it's good if we have more information and everything and then this is kind of
Excel like below.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 77 of 126
Project: KVH 2.0 AUD for KVH 2.0

They have the own format they will attach. So what basically for RDM what we need code and which is the
master code which will be mastered in EDW.
Then what is the name of that code and what we are saving Norwegian name and Norwegian description. So
you will have English name English description Norwegian name like that.
And Norwegian description. So if you can see that for one month like on frequency manner.
And another important column is sort order, If we want add any column so order will change.
we have some of these values are there ,those are like -1,-2,-3 but -3 is not using now a days ,it doesn't apply
when the RDM actually happens through inflow but -1 and -2 are important ,-1 is like whenever the null and
values getting from source and -2 is for unknown means some other values getting instead of actual .so RDM
cannot understand the those values(whenever they have defined unknown values from sources) .

Master tab : which is related to important master tab which have all the information about the code.
Meta tab:
This is also like updated by whoever is ordering the set so they will give the description of the code. What is the
Norwegian description effective date and Who is the owner of that particular set.

And we do not have any request or anything for sources and We only get request for master set here, so sets
and mappings will be updated in our mapping specification. I have been having specification and from there the
developer or the project team. Whoever is working on the project they took it from there and then they updated
in RDM.

RDM test: This is the link of RDM test.

It works under your error server.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 78 of 126
Project: KVH 2.0 AUD for KVH 2.0

This is the UI like below ss,and we have some different tabs in this main thing which that we want to look into
right now is this 01 master sets, so 01 master sets will have all your master sets which are mastered in EDW.

For example, if you see a AML_agreement country then you will see different versions of it. So how you create
like if you see here this AML agreement country might be first mastered in 20, 2020, 1906. So we give there is
an naming standard that we follow, which is a period and then underscore. 01 It starts with underscore 01
if you get multiple requests in the same period then it goes 02,03 and always have to see the approved version
of it. like below ss:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 79 of 126
Project: KVH 2.0 AUD for KVH 2.0

If you click on this you will have different sets, there are 368 values which are mastered for this AML agreement
country. And then if you click one of it then you will see that OK you have code name, description and there are
some other columns as well and you have the translations. This is the same Norwegian name and Norwegian
description so which we take from the accent sheet.

And we have one more dataset which is General set ,we have only one elimination of intercompany like below
ss:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 80 of 126
Project: KVH 2.0 AUD for KVH 2.0

This is the email we receive every month like below ss. We update it accordingly. As you can see, people from
FVCC are copied on it. This particular case involves the elimination of the Intel company. Every month, we
receive a request from this person—he updates some information, and then we have to retire it. This is an
exception case where he is the only one involved.

KVH2.0 sets:
We have some KVH and Housekeeping columns for which we perform lookups. For those, we also have two
sets of KVH data.

Source set:
specific sources related to our source data. As we know, we receive data from many sources, and each source
has its own folder.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 81 of 126
Project: KVH 2.0 AUD for KVH 2.0

whenever you create the source sets, we always give it the naming convention for follow the naming
convention like below this the master set name underscore source set code.

We have two kind of mappings those are copy mappings and independent mappings.
1. Copy Mappings: This seems to imply that the codes (or data) are being directly copied from the
source and inserted into the target without any change. If the data is already "mastered" in the system
(like KVH 2), then you're basically just copying the same values into the target, resulting in blanks
because no transformation or changes are needed.

2. Independent Mappings: These might involve some form of transformation, where the codes or values
from the source are mapped to the target in a way that is independent of the original source codes.
This could involve things like changing formats, applying business rules, or other transformations that
make the mapping not just a direct copy.

In the context of 84 ODS (which could be an Open Data Structure or a specific operation in a process),
So, when you mention that if the codes from the source are already mastered in KVH 2 and follow the same
structure, it would result in blanks, I assume you're saying that no transformation is needed, and hence there’s
nothing to fill out in the target mapping because everything aligns directly.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 82 of 126
Project: KVH 2.0 AUD for KVH 2.0

For more information refer below links:

Test RDM Link - IBM Reference Data Management Hub
RDM - New reference data orders

The IAP mapping specification sheet should be updated by your data modeler or data architect, depending on
who is working on it. for example,
We have worked on this Once updated below sheet, you will find an RDM tab in the mapping specification. This
tab will contain information on all the RDMs used for that particular mapping."

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 83 of 126
Project: KVH 2.0 AUD for KVH 2.0

If we see above ss, the duration type is being ordered.

If the mapping is defined as a copy mapping, the source codes are not provided. This means the values are
directly copied from the source to the target without transformation.
For source block status type or source record status type, the copy mapping is marked as no.
This indicates that you are receiving values from the source, but these values need to be mapped to specific
target codes. These target codes are usually referred to as master codes.
Essentially, instead of copying directly, you map the source values to the predefined master codes for
consistency and alignment.

RDM Lookups and Inflow Job:

RDM lookups are added as part of the inflow job, which is built using the Accelerator tool.

In the Ino utility, custom templates may be created for these RDM lookups. These templates are pre-
configured to handle the mappings or transformations.
Template Setup:
The key point is that you don't need to worry about how the RDM lookups are set up, as they have already
been configured in the system using the custom templates.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 84 of 126
Project: KVH 2.0 AUD for KVH 2.0

When you're working with these templates, it’s important to focus on creating the necessary datasets.
Dataset Name:
When creating a dataset, you need to ensure that the dataset name is correct.
The dataset name refers to the master set or target set—whatever name you’re using to define the dataset in
your system.

Inflow Generation and Mapping:

When we generate the inflow, the system runs a process that handles data based on the data provider code,
data source code, and the specific data provider chain and code that you have defined in your HIR (possibly
referring to a setup or configuration file) or IAP (possibly referring to a similar data mapping system).
Lookup Process:
The system uses the data provider code to identify which specific job is being run. Based on this, it performs a
lookup on the master set,which can be considered as the reference or target data set.
Mapping Validation:
Once the lookup is performed, the system checks the mapping for the provided data, ensuring that it matches
the predefined rules or values from the master set.
Job Execution:
The job will execute based on the mappings and lookups defined for the particular data provider code and the
associated codes and configurations within your setup.

The inflow job runs based on the configuration from the HIR/IAP, performing lookups on the master set and
validating against the mappings based on the data provider code.

->Business Location RDM Lookup:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 85 of 126
Project: KVH 2.0 AUD for KVH 2.0

The system will check the lookup for the business location RDM during the inflow job.
Based on this lookup, it will determine whether any values are missing or invalid in the data.
->Handling Null Values:
If the system finds null values in the lookup (for example, missing or invalid business location values), it will
map those nulls to default values.
As you mentioned, default values like -1 or -2 may be used to handle these nulls, ensuring that the job can
continue processing without errors.
->Error Logging:
If the job encounters an issue, such as being unable to find a specific RDM code, it will log the error.
The log will show:
The reason for the failure (e.g., missing data or incorrect mapping).
Which RDM code could not be found during the lookup.
The specific rows where the issue occurred, including the null values that were mapped to default codes.
->Debugging Failures:
If your job fails, you can refer to the log to see exactly what went wrong and which RDM code caused the issue.
This helps you identify any mapping problems or missing data that need to be addressed.
The lookup checks the business location RDM, handles null values by defaulting them to -1 or -2, and logs
any issues with specific RDM codes that couldn't be found, providing clear error messages for debugging.

Example below:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 86 of 126
Project: KVH 2.0 AUD for KVH 2.0

The system attempted to perform a lookup for agreement status_45364 with value 0 but couldn't find a
matching mapping in the RDM.
As a fallback, it returned -2 as the default value for this unmapped status code.

CD columns :
 Always have nullable = NO.

 Never contain NULL values.

 Even if the source sends NULL, it’s mapped to -1 by RDM utility.

Code Source columns :

 Always have nullable = YES.

 Can contain actual NULL values from the source.

 The reason is RDM (Reference Data Management) maps any incoming NULL in CD columns to -1
automatically.

 This ensures CD columns are always populated (never truly null), hence nullable = NO.

Deployment Check: whenever we are doing deployment checks

 After deploying a table, always verify

CD columns → nullable = NO
Code Source columns → nullable = YES
Because It’s a standard RDM check to maintain consistency and avoid nulls in CD fields.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 87 of 126
Project: KVH 2.0 AUD for KVH 2.0

DB2:

RDM & DB2 – Key Points:

1. RDM works via a UI-based application

o All operations (approve, update, etc.) are done through the UI.

2. If issues occur

o Contact the Platform Team if RDM or related services aren’t working.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 88 of 126
Project: KVH 2.0 AUD for KVH 2.0

o This applies especially when running jobs or updating value sets.

3. Direct DB updates not common

o In DMI (Data Management Interface), no manual updates are typically done on RDM tables.

o Any changes are expected to go through standard UI or platform processes.

4. RDM Tables reside in DB2 database

o The DB used is called MDMDB.

o Important RDM tables include:

 rd_value_set – contains the master value sets.

 MDMDB1.rd_value_set_relationship – stores relationships between value sets.

 Plus a few other supporting tables.

5. Helpful RDM query page exists

o There is a useful reference page with all relevant RDM queries (though the message cuts off
here).

If we want more info related to DB2 sql an tables, we can refer below confluence :
RDM - New reference data orders

RDM:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 89 of 126
Project: KVH 2.0 AUD for KVH 2.0

How to Create a New Master Set in RDM:

1. Receive Request

o Example: A change request comes in for AML agreement master set.

2. Search for Existing Set

o First, search the set in the current folder.

o If not found, check in other folders as well.

3. Create a New Version

o Go to the “New” option in the UI.

o Choose to create a new version of the master set.

4. Follow Naming Convention

o Use format: YYYYMM_<version>

o Example: 202504_01 for April 2025, first version.

5. Copy from Existing Version

o Copy all contents from the previous version.

o This allows you to add, update, or modify entries while keeping the base intact.

6. Save and Proceed

o Once created, continue making necessary changes in the new version.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 90 of 126
Project: KVH 2.0 AUD for KVH 2.0

What Happens After Creating the New Master Set Version:

1. Starts as a Draft Version

o The newly created master set initially is in "Draft" status.

2. Prepare an Excel Sheet for Updates

o Create an Excel file containing the changes or additions needed.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 91 of 126
Project: KVH 2.0 AUD for KVH 2.0

o This includes new entries or updates to existing records.

3. Import the Excel Sheet

o Go to the “Import” option in the RDM UI.

o Upload the Excel sheet to update the draft version.

4. Submit for Approval

o After import, perform the “Request Approval” action.

o This moves the draft version into the approval workflow.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 92 of 126
Project: KVH 2.0 AUD for KVH 2.0

Final Steps in RDM Master Set Lifecycle:

1. Submit for Approval

o After importing and reviewing changes, the set goes into "Pending Approval" status.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 93 of 126
Project: KVH 2.0 AUD for KVH 2.0

2. Approve the Set

o Go to “Action” and select “Approve”.

o This makes the version the new "Approved" master set.

3. Retire the Previous Version

o After approval, retire the older version of the master set.

o This ensures only one active version is in use at a time.

Full RDM Master Set Creation & Approval Process

1. Creating the New Master Set
 Go to the appropriate folder in the RDM UI.

 Click on "Create Set".

 Enter the necessary details:

o Name (e.g., 202504_01)

o Description (e.g., "Testing AML 2025 C")

o Type: Use M Default for single-key sets.

o Life Cycle: Select Default state machine.

2. Set Key Configurations

 For single-key master sets, M Default type is enough.

 For multi-key master sets, go to the "Administration" tab to define the additional keys.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 94 of 126
Project: KVH 2.0 AUD for KVH 2.0

3. Set Dates
 Effective Date: Current or required date.

 Expiration Date: Use 31-Dec-9999 to keep it active indefinitely.

 Review Date: Set to a date 2 days ahead (or per project standards).

 This helps the system recognize it as a test or temporary set, if needed.

4. Set Ownership & Metadata

 Define:

o Owner Logon ID (e.g., 62383)

o Physical Name (e.g., "TEST_ONLY")

o Code Name (e.g., "TEST_CD")

5. Import Data
 Prepare an Excel sheet with new or updated values.

 Use Import option in RDM UI to load the data into the Draft version of the set.

6. Submit for Approval

 Once the import is done and reviewed:

o Go to Action > Request Approval.

o This changes the status to "Pending Approval".

7. Approve the Set

 Authorized users go to Action > Approve.

 Once approved, the new version becomes active.

8. Retire Previous Version

 Retire the old version to avoid conflicts.

 Ensures that only one current version of the master set remains active.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 95 of 126
Project: KVH 2.0 AUD for KVH 2.0

Complete Steps to Create & Import a New Master Set in RDM

🔹 Step 1: Create the Master Set

1. Go to the correct folder in the RDM UI.

2. Click on "Create Set".

3. Fill in the following:

o Set Name (e.g., 2025_C)

o Description (e.g., "Testing AML 2025 C")

o Type: M Default (for single key). For multiple keys, configure them under the Administration
tab.

o Life Cycle: Use Default.

o State Machine: Also Default.

🔹 Step 2: Define Dates

4. Set:

o Effective Date – Current or as required.

o Expiration Date – Use 31-DEC-9999 (to keep it indefinitely valid).

o Review Date – Typically 2 days ahead of the current date.

🔹 Step 3: Add Ownership Metadata

5. Go to Set Properties and manually add:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 96 of 126
Project: KVH 2.0 AUD for KVH 2.0

o Owner Logon ID (e.g., 62383)

o Physical Name (e.g., "TEST_ONLY")

o Code Name (e.g., "TEST_CD")

🔹 Step 4: Import Data

6. Go to the “Values” tab — it will be empty initially.

7. Click Import, then:

o Choose your CSV/Excel file (e.g., from SharePoint).

o Map the columns:

 Code → code

 Name → name

 Description → description

 Review Date → review date

 Translation Values:

 Map to name Norwegian, description Norwegian, etc.

8. Preview the data, then click Finish.

o System will show a success message:

“28 records imported, 0 updated, no errors.”

🔹 Step 5: Final Steps

9. Verify imported data in the Values tab.

10. If everything looks correct:

 Go to Action > Request Approval

 After approval, this version becomes active.

11. Retire old version, if applicable.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 97 of 126
Project: KVH 2.0 AUD for KVH 2.0

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 98 of 126
Project: KVH 2.0 AUD for KVH 2.0

Finalizing RDM Master Set – Metadata & Value Import Steps

🔸 1. Set Expiration & Review Dates
 Expiration Date → Set to 31-DEC-9999 (keeps the record always active).

 Review Date → Set to 2 days ahead of the current date (used for testing/demo sets).

🔸 2. Set Ownership & Metadata (Manually)

 Go to Set Properties and manually enter:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 99 of 126
Project: KVH 2.0 AUD for KVH 2.0

o Owner Logon ID → e.g., 62383

o Norwegian Name → e.g., "Test Only"

o Physical Name → e.g., "TEST_ONLY"

o Code Name → e.g., "TEST_CD"

📝 These fields don’t appear by default, so you must enter them manually and then save.

🔸 3. Go to the Values Tab

 The Values tab will be empty initially for new sets.

 To populate it, you need to import values.

🔸 4. Prepare & Import Data from SharePoint

 Locate the CSV file (often shared via DM SharePoint).

 Example used: Duration Type Code.

 Even if the data is different, the import process remains the same.

🔸 5. Map the CSV Columns During Import

During the import:
 Map:

o Code → Code Column

o Name, Description, Review Date

o Translation Fields → Name Norwegian, Description Norwegian, etc.

 Preview the data on the next screen.

 Click Finish.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 100 of 126
Project: KVH 2.0 AUD for KVH 2.0

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 101 of 126
Project: KVH 2.0 AUD for KVH 2.0

Final Phase: Post-Import, Approval, and Retirement

🔸 1. Check Import Confirmation
 After importing the file:

o System message:
“Import is successful. 28 imports done. No updates. No errors.”

🔸 2. Verify Imported Values

 Click on the set → go to "Values" tab.

 Confirm that all expected values are present and correctly mapped.

🔸 3. Request Approval
 Once data is verified:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 102 of 126
Project: KVH 2.0 AUD for KVH 2.0

o Go to Action > Request Approval.

o The set status will change to "Pending Approval".

🔸 4. Approve the Set

 Then go to Action > Approve.

 Set becomes the active (current) version.

🔸 5. Retire Older Version (Optional)

 If a previous version of the master set exists:

o Retire it to avoid confusion or duplicate references.

o This keeps only the latest version in active use.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 103 of 126
Project: KVH 2.0 AUD for KVH 2.0

Creating a Source Set in RDM – Step-by-Step

✅ 1. Start from the Master Set
 After creating and approving the Master Set, the next step is to create a Source Set linked to it.

📂 2. Navigate to the Source Set Folder

 Go to the Source Set folder in RDM.

 This should be under the specific source system (e.g., E021).

➕ 3. Create the Source Set

 Click on "Set" > "Create".

 Choose the parent folder (e.g., E021) where the source set should be stored.

📝 4. Enter Source Set Details

 Set Name: e.g., TEST_E021

 Description:

o Add if available (e.g., "E021 Source Set for AML Agreement")

o If not, reuse the set name as description.

 Version Name: Follow the same convention as the master set

o Example: 202504_01 for April 2025, version 1.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 104 of 126
Project: KVH 2.0 AUD for KVH 2.0

Source Set Configuration – Additional Details

🔸 1. Type Selection
 Type: Set to SBS One Key

o Why? Because the source set only has one key (just like the master set).

🔸 2. Source Set & Mappings Are Editable

 Source set and its mappings are always set to “Active Editable”.

o This allows for ongoing updates as new data comes in from the source system.

🔸 3. Same Naming & Versioning Rules Apply

 Versioning follows the same format as master sets:

o Example: 202504_01

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 105 of 126
Project: KVH 2.0 AUD for KVH 2.0

Creating Independent Mappings in Source Set

🔸 1. Go to the Source Set
 Navigate to your active source set (e.g., TEST_E021).

🔸 2. Create a New Mapping

 Click on “New” to manually create a new value.

🔸 3. Fill in the Source Value Details

 Code: Enter the actual value received from source

o Example: One Month

 Name: If available, enter the display name.

o If not, just use the code as the name.

 Description:

o Add if known; otherwise, can be left same as name or blank.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 106 of 126
Project: KVH 2.0 AUD for KVH 2.0

Once the source set is updated with data from your source file, database, or HIR, go to the Mapping
section, click “New”, and select the source set for which you're creating the mapping to link it with the master
set.

To create a mapping in RDM, go to the Mapping section, click “New”, select the source set (e.g.,
TEST_E021) and target set (e.g., TEST_KITTY), then name the mapping using the convention
MAP_<SourceName>, set the version (e.g., 202504_01), choose the mapping type (Copy Mapping for direct
values or Source to Independent for manual mapping), and finally enter effective/expiration dates, description,
and save it — the mapping is now created and ready for manual value linking.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 107 of 126
Project: KVH 2.0 AUD for KVH 2.0

To manually map source values to master values, go to the created mapping, click “New”, then use the
option “Show only unmapped source values” to find values like One Month that aren’t yet mapped; select
the source value, choose the correct target (master) value, set the effective and expiration dates, and click
OK to complete the mapping.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 108 of 126
Project: KVH 2.0 AUD for KVH 2.0

RDM Updates – Roles, Responsibilities & Restrictions:

1. Not Everyone Handles It

o Only a few team members are responsible for RDM updates.

o Currently handled by Payal (onsite) and Kritika (offshore).

2. Why Limited Access?

o Because it’s a manual and sensitive process.

o Letting too many people make changes could lead to errors or inconsistencies.

3. Requests Come from Other Tech Teams

o These updates are often based on external requests, so careful handling is needed.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 109 of 126
Project: KVH 2.0 AUD for KVH 2.0

4. Authorized Personnel Only (in Production)

o Only designated people are allowed to create/update master sets in production.

5. Test Environment Is Okay for Practice

o Once you have access, you can try things in the test environment.

o But do not perform changes in production unless you're officially assigned.

4. LOGICAL AND PHYSICAL MODEL

4.1.ERWIN DATA MODEL

 Erwin is the tool used for all the development tasks, by implementing data
models. (uses 3rd NF)
 The current version of Erwin Data Modeler is 9.7 (migration planned to
v.12).
 It uses Vulcan Gitlab Model repository with Azure DevTestLabs VDI.
 The data models present are, CIM (logical), SDM (semantic), and PDM
(physical).
 ETL Pipelines are always CIM based and auto generated.
 Erwin modeler is a modeler which is used to visualize and understand the complex data
systems and it improves the data quality.
 Erwin is the tool we are using, so 9.7 is the version we are currently using but there is a
plan to upgrade to 11
 Erwin data model, it looks like that we have a different subject area and each subject
areas have different er diagrams.
 subject areas are like below, agreement on case, event finance, Group internal
organization and party

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 110 of 126
Project: KVH 2.0 AUD for KVH 2.0

If we take example party subject area it has four er diagrams like :

 Main display
 Privacy
 Common and resolution
 PAM

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 111 of 126
Project: KVH 2.0 AUD for KVH 2.0

If we take PAM small er diagram like entity relationship it is CMM(common information mode)l
and we should follow the some standards column name for the table like name should be the
capital and not under score and those has relationships like one -one and one to many and
hierarchy’s and data types.

And all the data types are predefined as per DNB because data types are categorized so we
should use those only.
we can navigate like below:

Erwin DMCm party Dnb information Model

And all the data types are pre-defined and categorized as per DNB, so we should use those standards .

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 112 of 126
Project: KVH 2.0 AUD for KVH 2.0

Design rules and model standards of Erwin Data Modeler:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 113 of 126
Project: KVH 2.0 AUD for KVH 2.0

4.2. DATA MODELS IN KVH 2.0

THERE ARE MAINLY FOUR DATA MODELS WHICH WE ARE USING AS

PART OF KVH 2.0:
1) cmm model
2) sdm
3) pdm
4) Kvh 9.5 model

SDM:
Sdm is a semantic data modeler, it is not a physical layer, but it is
kind of physical and which is the bridge between common
information model & physical model.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 114 of 126
Project: KVH 2.0 AUD for KVH 2.0

So here entities are represented different compare to CMM like

column name are mentioned with underscore and defined validity
and arrival time of the record and updated date and time.
And we can choose the keys and ordered and convert into SDM to
PDM and SDM is up to date.

PDM:

Physical data model is a modeler for generating the DDl

(metadata) for table creation

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 115 of 126
Project: KVH 2.0 AUD for KVH 2.0

And initially created the ddl by using pdm run it in Teradata and for
their future deploy ddl by using Automation Studio.

Mapping specification:
Mapping specification are different types such as
inflow specification,
inbound specification
outbound specification
reference data specification
source mapping specification etc..

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 116 of 126
Project: KVH 2.0 AUD for KVH 2.0

but usually used inbound flow and inbound specification.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 117 of 126
Project: KVH 2.0 AUD for KVH 2.0

If we take example of inflow specification:

Once data modeling completed the activity, update the information
and different kind of templet can use while generating the
influence specification.

And if we take example of agreement table of the influence job

specification above screen shot it has many subject areas defined
and entity’s are like delta templates and delta partition keys are
used in inbound flow speciation.

CIMX MODEL:
Cimx model is a Common model which will be generated, and it
can be used in two scenarios. One is new system comes, and we
want to send some data to data warehouse but multiple system
can use the same model. So by using the same model they can
send to us. So, there's something called CMM.

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 118 of 126
Project: KVH 2.0 AUD for KVH 2.0

And cimx model used different folders like baselines, Cimx

transformations and Cimx sub models, environment branches and
trainings

Baselines: base line is new setup but we not using this folders.
Cimx transformations :
Whenever we generate all the steps will be locally created And
these folders are used for the internal local runs.

sub models:
submodels configurations is there, but we are not using
environment branches
once CiMx is generated , it will have CMM schema and cmm CSV
file and HCR CSV file and different folder structures .

trainings:
it created for couple of Mark runs. like kind of a full file run and if
we want to remove the particular attribute or if we modify the

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 119 of 126
Project: KVH 2.0 AUD for KVH 2.0

attribute it will impact everywhere because those connected

different entities so we should use training before that but it not in
use now a days.

5. REFERENCES AND DOCUMENTS

5.1. REFERENCES

Home Page for KVH2.0 :

 This link includes all the topics about KVH2.0

EDM - KVH2.0 - EDW Service Delivery - DNB Confluence

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 120 of 126
Project: KVH 2.0 AUD for KVH 2.0

 Monthly run is called as Kort Run and the link is attached below(will be covered by Simran
and kirthika):
KVH2.0 Monthly Kort Run - EDW Service Delivery - DNB Confluence

 The 1219 page has all the necessary topics like DI Hub and target tables, etc and
attached below:
1219 DI Hub - DNB Confluence

 BI IM page for old technical fields and aspects:

BI IM - DNB Confluence

 SDM to PDM step by step process by Erwin Model:

Table creation using Erwin - DAP Teradata - DNB Confluence

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 121 of 126
Project: KVH 2.0 AUD for KVH 2.0

 HIR flow generation:

Inbound Flow generation How-To - 1219 DI Hub - DNB Confluence

 Smart Admin old reference pages:

1300 KVH - Maintenance Activity - Infosys BI AO - DNB Confluence

 CMM Guide:
CMM-Guide - IT Business Intelligence - DNB Confluence

 SharePoint for Mapping Specifications:

1219 DI Hub - Specifications - All Documents

 CMM Guide:
CMM-Guide - IT Business Intelligence - DNB Confluence

 Link for CMM File Specifications:

CMMFile Specification - 1219 DI Hub - DNB Confluence

 Link for DIHub HR:

DIHUB HR - 1219 DI Hub - DNB Confluence

 DIHub Source (Finance ODS):

DI Hub Source: 1602 Finance ODS - 1219 DI Hub - DNB Confluence

 RDM Confluence Link:

https://dnb-asa.atlassian.net/wiki/spaces/RDM/overview?homepageId=130449410

Mapping Specifications sheet:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 122 of 126
Project: KVH 2.0 AUD for KVH 2.0

 KT Videos sessions:
Enterprise Data Management - KVH2 - All Documents

 GitLab link:
Working with Gitlab - EDW Service Delivery - DNB Confluence

 Onboarding New Members in DNB:

Onboarding New Member - EDW Service Delivery - DNB Confluence

Shared drives:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 123 of 126
Project: KVH 2.0 AUD for KVH 2.0

Drive for any Software, created by DAP team:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 124 of 126
Project: KVH 2.0 AUD for KVH 2.0

VedApp mostly used by KVH1, but needed for KVH2.0 sometimes, They are used for
batch run:

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 125 of 126
Project: KVH 2.0 AUD for KVH 2.0

5.2.DOCUMENTS

© 2025 Capgemini and DNB Confidential - All rights reserved

Page 126 of 126