You are on page 1of 26

Efficient usage of

BW InfoProviders
Product Management SAP NetWeaver / BI
SAP NetWeaver Regional Implementation Group - BI
SAP AG
Contents

Contents

Overview
Analyzing business and technical requirements
Using ODS Objects the right way
Modeling data flows
BW InfoProviders with precalculated data
Efficient usage of BW MultiProviders

SAP AG 2004, Efficient usage of BW InfoProviders / 2


DW Layers in SAP BW

Data Warehouse Operational Data Store Multidimensional Models


Non volatile Operational Reporting Multidimensional analysis
Granular Near Real-Time / Volatile Aggregated view
Integrated Granular Integrated
Historical foundation Built with ODS Objects Built with InfoCubes
Built with ODS Objects

... to provide the right information for all users


SAP AG 2004, Efficient usage of BW InfoProviders / 3
InfoProvider in SAP BW: Overview

Business
Explorer
Basic InfoCube
Data Contained

InfoProvider Interface
InfoObjects

OLAP
Engine
MultiProviders
ODS Object

Virtual InfoCubes

InfoSet

SAP AG 2004, Efficient usage of BW InfoProviders / 4


Scope: Efficient usage of BW InfoProviders

Project

Project

Project

Building your DW Modeling (single)


and EDW strategy BW InfoProviders

In between these two concepts you have to decide on:

What BW InfoProvider types to use

How many BW InfoProviders you will use

How to model these BW InfoProviders in data flows

When and how to use BW MultiProviders

SAP AG 2004, Efficient usage of BW InfoProviders / 5


Contents

Contents

Overview
Analyzing business and technical requirements
Using ODS Objects the right way
Modeling data flows
BW InfoProviders with precalculated data
Efficient usage of BW MultiProviders

SAP AG 2004, Efficient usage of BW InfoProviders / 6


Analyzing business and technical requirements I

The following selected areas from your business blueprint have to


play an important role when deciding on BW InfoProviders:
Granularity
How do you want to look at the data? Daily? Monthly? Quarterly?
What level of analysis needs to be done? Aggregated or detailed?
Are snapshots of the data needed? Does data need to be restated?

Data Volumes
How many line items are there? How many products do we have?
How many years of data do we need to house? What data history do
you want to keep in your data warehouse ?

Analysis aspects
How is data being analysed? OLAP analysis including navigation on
data or flat list reporting?
What response times do my users expect to return their results? How
quickly do they need access to the data?

SAP AG 2004, Efficient usage of BW InfoProviders / 7


Analyzing business and technical requirements II

Data Sources
How much transformation of the data is required? Does this need to
take place at multiple times during the data loading process? Is the
data going to be reported on throughout the transformation process?
Structure of source data: Does the source system send normalized
(document like) or already pre-aggregated data? Is delta calculation
required on source data?

Knowing about your business requirements and your technical


environment (e.g. data volumes and data sources) is the basis for
all high level and low level modeling in SAP BW.

The quality of the implemented data models and data flows is the
basis for all administration and performance tuning in SAP BW.

SAP AG 2004, Efficient usage of BW InfoProviders / 8


Contents

Contents

Overview
Analyzing business and technical requirements
Using ODS Objects the right way
Modeling data flows
BW InfoProviders with precalculated data
Efficient usage of BW MultiProviders

SAP AG 2004, Efficient usage of BW InfoProviders / 9


Using ODS Objects the right way

Check your business requirements against ODS Object features:


ODS objects serve to store consolidated
and cleansed data on a document level from
one or more InfoSources. Usage examples:
Updating further data targets
Data validation
Data in an ODS object can be analyzed by
the Business Explorer mainly to support
operational reporting.
ODS Objects can support calculations on
document level or calculation of delta on
source data
Think of deleting (calculated) delta
records from your change log table
that are not needed anymore.
ODS Objects can support central master
data consolidation

SAP AG 2004, Efficient usage of BW InfoProviders / 10


Using ODS Objects bad example

InfoCube
Cost Centre
Controlling
Redundant storage of
data on same level of
granularity
Additional benefit
of second
No transformations ODS Object?

ODS Object: ODS Object:


Cost Centre Controlling Cost Centre Controlling
Actual Data Plan Data

No document like
data, no delta
calculation required
Customer Data Source delivering delta records

SAP AG 2004, Efficient usage of BW InfoProviders / 11


Using ODS Objects good example

Data Mart for OLAP InfoCube:


analysis on Sales analysis
aggregated data
Calculations on
document level

Flat List Reporting


on consolidated ODS Object:
Document flow in sales & distribution
data

Build DWH
layer that
stores ODS Object: ODS Object:
document Sales Orders Deliveries
data for Item Level Item Level
reference

SAP AG 2004, Efficient usage of BW InfoProviders / 12


Contents

Contents

Overview
Analyzing business and technical requirements
Using ODS Objects the right way
Modeling data flows
BW InfoProviders with precalculated data
Efficient usage of BW MultiProviders

SAP AG 2004, Efficient usage of BW InfoProviders / 13


Modeling data flows

Design your data flows according to


business and technical requirements.
This has to include:

Knowing about your source data and


the needed levels of granularity in your
BW data flow

Deciding on what level of granularity

/
transformations have to occur

Avoiding redundant data storage in


order to optimize disk space

Reducing complexity of your data flow


for improved performance and
administration

Implementing data validation


procedures where needed

KISS - Keep it small and simple!


SAP AG 2004, Efficient usage of BW InfoProviders / 14
Example on data flow modeling - before

Only transformations InfoCube:


on material level. No Sales analysis
transformation on
sales order item level.

InfoCube:
All DataSource / PSA Sales analysis
information is kept in
the ODS Object

ODS Object:
Sales Orders
Item Level

Reloading data only


occurs on very last
requests. Building or
Rebuilding InfoCubes is
PSA for DataSource
Sales Order Items
done based on ODS Object
and Source system
ERPCLNT100
SAP AG 2004, Efficient usage of BW InfoProviders / 15
Example on data flow modeling - after

If no transformations Possible modeling as


occur between ODS aggregate of other
If transformations Object and InfoCube, InfoCube as nearly on
occur, this
they can be loaded in same level of detail
pragmatic approach
might not be parallel
chosen in order to
keep single point
of truth
InfoCube:
Sales analysis

ODS Object:
Sales Orders
Item Level

ODS Object might respond to


all administrative needs.
Another option would be to
use PSA, but to keep only
PSA for DataSource
Sales Order Items
very recent data in it.
and Source system
ERPCLNT100
SAP AG 2004, Efficient usage of BW InfoProviders / 16
Contents

Contents

Overview
Analyzing business and technical requirements
Using ODS Objects the right way
Modeling data flows
BW InfoProviders with precalculated data
Efficient usage of BW MultiProviders

SAP AG 2004, Efficient usage of BW InfoProviders / 17


BW InfoProviders with precalculated data

Query Design and Scope of this Potential for the


OLAP Analysis presentation Information Consumer

A BW InfoProvider Exceptionally, a BW BW Information


contains transformed InfoProvider can contain Consumers can receive
and integrated data for precalculated, query like precalculated query
multiple OLAP analysis data in order to support documents using BEx
advanced analysis Broadcaster

Degree of Precalculation
SAP AG 2004, Efficient usage of BW InfoProviders / 18
Example 1: Inventory Management and Snapshots

Snapshot scenarios can be an alternative to non cumulative key


figures and non-cumulative InfoCubes if:
High number of movements on object for which the stock has to be
calculated = Stock quantity changes frequently
High number of objects for which a stock quantity has to be calculated
Frequent analysis using averages or historic stock values

InfoCube: InfoCube:
Actual stock quantity Stock quantity on material
on material level level (end of month)

Update of month end snapshot


Daily refresh (full upload) once a month

Stock is kept and ODS Object:


calculated in ODS Actual stock quantity
on material level
Object, not in OLAP

Opening balance Material movements

SAP AG 2004, Efficient usage of BW InfoProviders / 19


Example 2: Analysis Process Designer

The Analysis Process Designer allows you to


Access and transform information using a single data process
combining different sources and different analysis steps

Simple Example: Send query results to a


transactional ODS Object (and from there to
further BW InfoProviders in your data flow)

Deploy key performance indicators and analytical results to certain


BW InfoProviders or to your operational processes in SAP CRM

SAP AG 2004, Efficient usage of BW InfoProviders / 20


Contents

Contents

Overview
Analyzing business and technical requirements
Using ODS Objects the right way
Modeling data flows
BW InfoProviders with precalculated data
Efficient usage of BW MultiProviders

SAP AG 2004, Efficient usage of BW InfoProviders / 21


Efficient usage of BW MultiProviders I

BW MultiProviders are used for


Combining and analyzing heterogeneous business data (different key
figures) out of different BW InfoProviders using common characteristics

Sales Costs Forecast

Splitting data of one business scenario to separate Basic InfoCubes


(Logical Partitioning)
Possible partitioning criteria: year, plan/actual, regions, business area
Use MultiProviders to cut large amounts of data into chunks

2002 2003 2004

SAP AG 2004, Efficient usage of BW InfoProviders / 22


Efficient usage of BW MultiProviders II

Benefits
No additional data storage
Single InfoProviders smaller, less complex and less sparsely filled
than one big InfoProvider
(Parallel) Data load into individual InfoProviders
Better performance in administration on Basic InfoProviders
In most cases, queries are split automatically and distributed to
InfoProviders (for more details on parallel execution see SAP Service
Marketplace note 629541)
Central interface for querying even if underlying data models change

Disadvantages
Administration (with aggregates)
Increasing number of InfoProviders to be administrated

SAP AG 2004, Efficient usage of BW InfoProviders / 23


Efficient usage of BW MultiProviders III

Make sure that your BW MultiProviders only retrieve data out of


relevant InfoProviders at query runtime.

By using constants in the By using


design of Basic characteristic
InfoProviders (to be 0INFOPROV when
specified at implementation Multi
Multi designing a query
time) Provider
Provider on the
MultiProvider.
Example:
0INFOPROV =
ZEUSALES

By using
different key
0REGION = 0REGION = 0REGION = 0REGION =
figures in your EMEA EMEA AMERICAS APA
Basic
InfoProviders Asia
Europe Europe America Pacific
Inventory Sales Sales Sales
0QUANTITY 0AMOUNT 0AMOUNT 0AMOUNT

SAP AG 2004, Efficient usage of BW InfoProviders / 24


Further Information

Public Web:
www.sap.com > Solutions > SAP NetWeaver

SAP Service Marketplace:


http://service.sap.com/bw
z BW InfoIndex ODS Functions
z BW InfoIndex Enterprise Data Warehousing
z BW InfoIndex Data Modeling
z BW InfoIndex Non-cumulatives
z Folder Performance

http://service.sap.com/education
z BW330 BW Modeling (3 days)
z BW360 BW Performance & Administration (5 days)

SAP AG 2004, Efficient usage of BW InfoProviders / 25


Copyright 2003 SAP AG. All Rights Reserved

No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of
SAP AG. The information contained herein may be changed without prior notice.

Some software products marketed by SAP AG and its distributors contain proprietary software components of other software
vendors.

Microsoft, WINDOWS, NT, EXCEL, Word, PowerPoint and SQL Server are registered trademarks of
Microsoft Corporation.

IBM, DB2, DB2 Universal Database, OS/2, Parallel Sysplex, MVS/ESA, AIX, S/390, AS/400, OS/390, OS/400, iSeries,
pSeries, xSeries, zSeries, z/OS, AFP, Intelligent Miner, WebSphere, Netfinity, Tivoli, Informix and Informix Dynamic
ServerTM are trademarks of IBM Corporation in USA and/or other countries.

ORACLE is a registered trademark of ORACLE Corporation.

UNIX, X/Open, OSF/1, and Motif are registered trademarks of the Open Group.

Citrix, the Citrix logo, ICA, Program Neighborhood, MetaFrame, WinFrame, VideoFrame, MultiWin and other Citrix
product names referenced herein are trademarks of Citrix Systems, Inc.

HTML, DHTML, XML, XHTML are trademarks or registered trademarks of W3C, World Wide Web Consortium, Massachusetts
Institute of Technology.

JAVA is a registered trademark of Sun Microsystems, Inc.

JAVASCRIPT is a registered trademark of Sun Microsystems, Inc., used under license for technology invented and implemented
by Netscape.

MarketSet and Enterprise Buyer are jointly owned trademarks of SAP AG and Commerce One.

SAP, SAP Logo, R/2, R/3, mySAP, mySAP.com and other SAP products and services mentioned herein as well as their respective
logos are trademarks or registered trademarks of SAP AG in Germany and in several other countries all over the world. All other
product and service names mentioned are trademarks of their respective companies.

SAP AG 2004, Efficient usage of BW InfoProviders / 26