Professional Documents
Culture Documents
Enterprise Data Warehouse (EDW) With SAP BW
Enterprise Data Warehouse (EDW) With SAP BW
Content
Content Enterprise Wide BW - Background Elements of a Corporate BW Strategy The BW Data Model Prevents Silos BW Data Layers Guarantee Reliability and Flexibility BW Corporate Landscape Patterns
Enterprise Wide BW
Background Continuously increasing number of BW installations Steady growth and coverage of BW installations High degree of customer-satisfaction with BW Broad analyst acceptance of BW and the packaged solution approach In general a renaissance of data warehousing
Introduce an enterprise wide BW Strategy and Guidelines that guarantees : Flexibility Reliability at low costs What offers BW to help you realizing an enterprise Data Warehouse Strategy ? What are the elements of an enterprise BW Strategy ?
SAP AG 2004, Enterprise Wide BW, J. Haupt / 3
Enterprise Perspective on BW
Enterprise perspective on BW: Higher level organizational units want to harmonize the lower level BW initiatives Architecture-based BW implementation from the very beginning Stabilize a mature BW landscape Division A Division A (Overhaul the architecture)
...... ......
Group Group
Division B Division B Division C Division C Division D Division D
Retail Retail
....... .......
Source
Unit 1 BW
Consistency issues: Uncontrolled data flows Source Source Multiple extractions of same data Costly development Local BW Not aligned data models Source
Stream C
Source
Local BW
BW Unit 2 BW
Source
....
BW Project Team
Potential consequences: Departmental islands (silos) Consistency problems Completeness not sufficient Flexibility restricted Trace back not possible
If there is no organizational momentum toward a common goal, then the best architecture, the best framework in the world is bound to fail.
W.H. Inmon
Who needs a Corporate BW Strategy? All, but the prerequisites are different!
Group Headquarter Stream A Stream B Stream C Stream D
Organization, Business
...... Retail Business Marketing ......
countries
Regions/ countries
Legacy
OLTP landscape
Legacy
Legacy R/3
Content
Content Enterprise Wide BW - Background Elements of a Corporate BW Strategy The BW Data Model Prevents Silos BW Data Layers Guarantee Reliability and Flexibility BW Corporate Landscape Patterns
Landscape
Operational Decision-Making
?
A challenge for Data Warehouse consistency
Sales OLTP
HR OLTP
BW Data Warehouse Data Model for SAP und other applications (InfoObjects & InfoSources)
BW InfoCube:
scope-specific
BCT InfoSources 0MATERIAL 0MATGR 0MATTYPE 0DOCNO 0CUSTOMER 0SALESPERS 0MATERIAL 0AMOUNT Subject Areas
Master Data
Transaction Data
Customer
Sales Person
Local Part of
Material Dimension
0MATERIAL
Material_Dimension_ID 0MATERIAL Material Dimension Table
Material_Dimension_ID SalesOrg_Dimension_ID Time_Dimension_ID Customer_Dimension_ID Sales Amount Quantity
Material Type
0MATERIAL
Language Code Material Name
Region 2
Region 3
InfoCube A
Bezirk 1
Bezirk 2
Bezirk 3
Bezirk 4
Bezirk 5
Key figure 1
FACT Table
Between people
Enterprise BW Strategy:
Usage of delivered BW DWH data model whenever possible Central controlled guidelines for modifications and expansions
Content
Content Enterprise Wide BW - Background Elements of a Corporate BW Strategy The BW Data Model Prevents Silos BW Data Layers Guarantee Reliability and Flexibility BW Corporate Landscape Patterns
SAP Business Information Warehouse and the Corporate Information Factory (CIF)*
Data Acquisition
Staging Area Cleansing Transform
Data Delivery
Architected Data Marts
Finance
any source
Open D is tribu ti on
Transaction Data
Logistic
The Data Mart is customized and/or summarized data derived from the data warehouse
any target
BW Architected Data Mart Layer: The Departmental Layer What can we expect?
0CUSTOMER master
0CUSTOMER InfoCube B
0MATERIAL master
The BW Data Mart Layer based on the BW Data Model and a proper Project Method offers
Consistency within the Data Marts (InfoCubes, ODS-Objekte) Control of data redundancy (MultiProvider, Queries) All desired historical views ...
SAP AG 2004, Enterprise Wide BW, J. Haupt / 18
SAP Business Information Warehouse and the Corporate Information Factory (CIF)*
Data Acquisition
Staging Area Cleansing Transform
Data Deliver
Architected Data Marts
Open D is tribu ti on
Finance
any source
Transaction Data
Logistic The result of the data transformation and cleansing process is stored persistently: Subject oriented Integrated Granular * Source: Bill Non volatile (historical) Inmon Not scope-specific (not flavored)
any target
The BW (Enterprise) Data Warehouse Layer The Corporate Layer What can we expect?
Reliability, Trace back Prevent Silos
Single point of truth All data have to pass this layer on its path from the source to the summarized edw-managed data marts
Integration
Data are integrated (as far as possible) Realization of the corporate data integration strategy
...
The BW Data Warehouse Layer is the corporate memory, the corporate information repository
SAP AG 2004, Enterprise Wide BW, J. Haupt / 21
Project Team
1
Set up
Incremental set up in parallel with data marts guarantees buy in from business users
2 2 1
Business Logic
1 1 1 1
(E)DW Layer
(ODS-Objects) (ODS-
Technology
BW ODS-Objects Expanded InfoSources define ODS-Object structures BW flexible staging for Master Data
SAP AG 2004, Enterprise Wide BW, J. Haupt / 22
1 Increment
SAP Business Information Warehouse and the Corporate Information Factory (CIF)*
Data Acquisition
Staging Area Cleansing Transform
Data Deliver
Architected Data Marts
Open D is tribu ti on
Finance
any source
Transaction Data
Logistic
any target
Operational, Tactical and Strategic Decision-Making in Traditional BW Implementations Most BW customers do not distinguish between strategic/ tactical and operational decision-making using BW features:
Data Marts
Standard Collections
InfoCube with precalculated aggregates Two InfoCubes with Drill Thru InfoCube and ODS-Object with Drill Thru
ODS-Object
standard collections
detailed, slightly summarized InfoCubes, ODS-Objects
Limits of Using BW the Traditional Way: Introducing BW Operational Data Stores The traditional usage of BW (with or without dwh-layer), which supports all kinds of decision-making using Architected Data Marts has its limits
If we want to provide event-level or close to event-level operational reporting (near real time data warehousing) and/ or If we are confronted with high data volumes (e.g. Retail)
BW Operational Data Store
I no f Cub e O S D O j ect b G ran l a X u rity
I no f Cub e
Data Marts
G ran l a > X u rity
DWH
G ran l a >=X u rity
G ran l a X u rity
Dedicated ODS-Objects (or InfoCubes) with granular data support operational decision making Directly loaded from PSA Short life cycle of data ( Archiving, Near line Storage) They build a BW Operational Data Store
G ran l a X u rity
Abstract Layer Architecture and the Reality of BW as Packaged Solution Focus of traditional BW implementations:
tactical / strategic
decision-making
decision-making
Content
Content Enterprise Wide BW - Background Elements of a Corporate BW Strategy The BW Data Model Prevents Silos BW Data Layers Guarantee Reliability and Flexibility BW Corporate Landscape Patterns
Changed Data
eComm.
EDW
Global ODS
Oper. Mart
Bus. Int.
local ODS
Granularity Manager
Archives
Web Logs
Internet
SAP AG 2004, Enterprise Wide BW, J. Haupt / 29
Local
BW
Local
BW
Local
Global BW
BW
Select the [IT] approach that fits most closely with corporate Prof. Sethi, University of Texas strategy
Information&Management, 2/01
Local interests
local freedom central governance
External Data Marts BW Architected Data Marts Data Standard Marts Collections latency: >= 1 day mission critical business critical others
Staging Area
BW EDW
BW ODS
BW EDW
BW ODS
ODS
Global BW
Parameters Scalable Mission critical dominates EDW EDW separated from majority of data marts
SAP AG 2004, Enterprise Wide BW, J. Haupt / 34
ODS
ODS
Summary
BW offers a wide range of features that support enterprise wide data warehousing needs. Nevertheless a strategy and an organizational momentum must exist to realize an enterprise wide BW strategy.
Questions
Any Questions?
Further Information
Public Web:
www.sap.com > Solutions > SAP NetWeaver
SAPNet:
Use ALIAS: /BW
www.service.sap.com/education
PDEBW1 Enterprise Wide Data Warehousing with SAP BW (Workshop - 2 days)
SAP AG 2004, Enterprise Wide BW, J. Haupt / 37