Kalido: Data Warehouse Lifecycle Management

1
Kalido Data Warehouse Lifecycle Management
2
Kalido Data Warehouse Lifecycle Management
(DWLM)

• Kalido is a software company that provides a
  solution that automates and integrates the
  design, development, deployment, maintenance and
  management of data warehouses and data marts.

3
Need to align IT with the Business!!!
IT Cycle
Business Cycle
(Data Warehousing)
Prepare
Operate
Plan
Evaluate
Run
Implement
4
Traditional DW are custom-built and maintained
Reporting
Sources
Design / Build
ETL
ETL
Any change in business requirements potentially
means

• Changes to the database structure
• New load routines and transforms

Time Cost Effort Risk

• New validation programs
• New summarization programs

• New performance optimization
• Reconfiguring of reporting tools

5
Physical World - Manufacturing / Distribution
Accounts
Facilities and Equipment
Products
Strategic Sourcing
Stocks and Shares
Sell
Competitors
Negotiation
Channels
Enterprise Business Model
Deliver the Products
Buy
Suppliers
Customers
Define Manage Supply Chain Service
Sales
Manage Operate the Site
Calendars
Production
Staff
Organizations
Locations
6
Business model the common language

• Reflects the business structure and relationships
• Representation of how the business is organized
• Changes all the time
• Restructuring, MA, new data sources
• Inherently complex
• Multiple perspectives marketing, finance,
  operations, 3rd party
• Business model IT representation
• Must be reflected logically and physically in the
  data warehouse and data marts
• Must be flexible to handle change

Organization
7
The IT Representation of the physical world
8
KALIDO DWLM Architecture
Source System, ETL, EAI
Business Model
Meta Data
Data
Meta Reference Data
Meta Data
Create Star Schema
Generate Dimension Tables
Raw Data
Data Marts
Loading Validation
Queries
Warehouse (Star Schema)

• File
• ODBC
• FTP

• Table
• View
• Text
• Excel
• Cube

Direct Access
9
Kalido Data Definitions

• gtgt Meta Data Characteristics
• Volatile
• Very Low Volume
• gtgt Reference Data Characteristics
• Highly Volatile
• Low Volume
• Basis for integration
• Foundation for rich analytics
• Key to managing change
• gtgt Transaction Data Characteristics
• Stable
• High Volume

Meta Data
Reference Data
Transaction Data
gtgt Meta and Reference Data are managed in
Kalidos business modeling Data Store.
10
Optimized for Maintenance and Reporting
gtgt Logical Model
gtgt STAR SCHEMA
gtgt Automatic and Highly Optimized Generation and
Maintenance of Type-2 and Hybrid Slowly Changing
Dimensions

11
Powerful Modeling Capability
Product
Customer

Brand Manager
Product Segment
Product Class
Industry
Industry Class
Product Group
Brand
Site Facility
Industry Sub Class
Product
Commercial Product
Retail Product
Customer

• Many to many
• Super / sub type
• Self referencing

• ERP Sales
• Volume
• Revenue
• Cost

• Legacy Sales
• Volume
• Revenue

12
Merger Semantic Integration
COMPANY B
COMPANY A
Business Unit
Subsidiary
Business Unit
Region
Department
Sales Office
Sales Rep
Sales Rep
gtgt FLEXIBLE REFERENCE DATA FORM A POWERFUL BASIS
FOR INTEGRATION
Transactions
Transactions
13
Internal Data Architecture
Meta and Reference Data
Transactions
Data Marts
Aggregates
gtgt OPEN DATABASE SCHEMA IN ORACLE, SQL SERVER, OR
DB2
14
KALIDO DWLM suite fits into your architecture
Kalido Direct Extraction
Data Warehouse Lifecycle Management
Source Systems
ETL

• Excel
• Business Objects
• Analysis Services
• Cognos
• Hyperion
• DB2 OLAP Server

• JMS
• MQ / Tibco
• Informatica
• Ascential
• Acta

• I2
• SAP
• Oracle
• JD Edwards
• Peoplesoft
• Siebel

• DB2
• SQL Server
• Oracle

Databases
Go to one place to manage change!

• HP
• IBM
• Sun
• Microsoft

Platforms
15
KALIDO DWLM- augmenting a mature environment
New Data
New Data
POS
Enterprise Data Warehouse
Data Warehouse Lifecycle Management
Delivery
ERP

• Excel
• Business Objects
• Analysis Services
• Cognos
• Hyperion
• DB2 OLAP Server
• Etc.

• DB2
• SQL Server
• Oracle

Legacy
Databases

• IBM
• HP
• Sun
• Microsoft

Platforms
DWLM can accelerate the data delivery process by
building on the existing DW infrastructure
16
KALIDO DWLM ROI v Custom Build

gt 48 saving on design through test gt 55
saving on maintenance
Source META Group-audited study
17
Warehouse lifecycles

• With custom-built data warehouse implementations
• Data warehouse cycles lag behind business cycles
• Backlog of requirements build up
• Business agility reduced
• Data warehouse maintenance costs increase over
  time
• Opportunity cost
• Data mart anarchy
• Big bang (12 18 months)
• With Kalido DWLM
• Business Cycle Data Warehouse Data Mart Cycle
• Reduced TCO
• Improved top line growth
• Iterative (3-4 months)

18
GANTT charts of merger acquisition
change(independently audited study)
Custom-Built Data Warehouse
KALIDO Data Warehouse
19
Kalido DWLM - whats in it for IT???

• Strategic deliverables vs. fighting fires
• Job security
• Automate the mundane focus on strategic user
  requirements
• Get ahead of the business cycle
• Spend time modeling the business vs. developing
  custom code
• Top line growth tied to IT deliverable
• Sound business management based on reality
• Faster time to market eliminate backlog
• Price tag on a good decision vs. a bad decision

20
Company background

• Spin-out of Royal Dutch Shell
• Funded by Atlas Venture and Benchmark Capital
• Award-winning, patented, KALIDO DWLM Suite
  implemented in over 250 locations in over 100
  countries
• 125 person years RD investment
• Strategic alliances and technology partnerships
• Ascential, IBM, Accenture, BO
• 100 of our customers said they would recommend
  DWLM to others (Kalido survey 2003)
• Headquartered in Burlington, Mass, and London
  with operations in US, UK and France

21
Customer Case Studies KALIDO DWLM in Action
22
(No Transcript)
23
Unilever Universal Information Project

• Project
• Integrate information on global brands to
  support Path to Growth strategy
• Consolidate reference data
• Transaction feeds from SAP / legacy
• Staged roll-out by world wide
• Benefits
• Better brand data delivered direct to brand
  managers worldwide
• Improved supply chain performance
• Adapt to MA Activity
• Visibility across organization
• Corporate memory

24
Shell Oil Products Management Information System
(OPMIS)

• Project
• Global data warehouse
• SAP, JDE and SunSystems ERP
• Core warehouse model
• Central reference data base
• Federation rollout
• Zonal consolidation process
• Benefits
• Centralized management model
• Consistent performance reporting
• Margin improvement
• Single view of customers and products
• Speed of new business integration
• Completed on time (80 countries in 1.5 years)
  and within budget