Extending the Reach of LSA++ Using

New SAP BW 7.40 Artifacts

Pravin Gupta, TekLink International Inc.
Bhanu Gupta, Molex
SESSION CODE: BI2241

Agenda

What is Enterprise Data Warehousing (EDW)?

Introduction to Layered Scalable Architecture (LSA)

Migration from LSA to LSA++

New SAP BW 7.40 Modeling and Provisioning Artifacts

Operational Data Provisioning (ODP) and Operational Data Queue (ODQ)

Open ODS View
Advance DSO
Composite Provider
Automatic HANA View Generation

Actual Use Cases

Conclusion

Why Enterprise Data Warehousing (EDW)

Consolidate the data across the enterprise to get a consistent and
agreed view on your data
"Having data is a waste of time when you can't agree on an interpretation.

EDW requires a database + X

SAP BW (BW) is the X as EDW with BW provides a flexible and
scalable EDW solution
Highly integrated tools for modeling, monitoring and managing the EDW
Open for SAP and non-SAP systems

EDW vs. Datamarts

Layered Scalable Architecture

LSA is the accepted approach for building EDW guaranteeing a consistent,
highly available and maintainable data foundation for an agreed interpretation of
BI & Reporting
Architected
data mart
layer

Visualization Layer

Enterprise
Data
Warehous
e Layer

Data Propagation Layer

Quality & Harmonization Layer

Corporate
Memory

LSA

Business transformation Layer

Operational
Data Store

Reporting layer (Architected Data Marts)

Data Acquisition Layer

Source: SAP

In-Memory Evolution

Enterprise Data Warehouse and Data Mart

Modeling with SAP Net Weavers BW

Data
Modeling

Data
Persistency at
Runtime

BW 7.0
DB + BWA 7.0

BW 7.3
DB + BWA 7.2

BW 7.4 from
SP5
on HANA DB
In Memory
planning engine

Planning Engine

Analytic Engine

First calculation
scenarios in BWA

Additional
calculations
in-memory

Data Manager

Filter + aggregation

Multi provider
handling and
flexible joins

Consumption of
HANA model in BW

InfoCubes

BWA instead of
aggregates

BWA-only
Info Cubes

DataStore Objects

EDW Processes

BWA reporting
for DSOs

SAP HANA
optimized Info
Cubes
SAP HANA
optimized DataStore
Object

HANA data for

BW staging

Source: SAP

LSA++ Holistic Framework

BI Streamlined: On EDW, Operational BI, Agile BI, Virtualization

Source: SAP

Journey From LSA to LSA++

Migrate to Hana- Optimized objects

Streamline EDW Core

Enhance Virtualization Layer

Introduce additional layers

Open Operational Data Store

Agile Data Mart Layer

BW Workspace Layer

Resulting LSA++

LSA vs. LSA++

LSA

No reporting on Propagator.
Result of transformations stored in additional persistent Layer, i.e. Architected Data
Mart.
Virtualization Layer only on top of Architected Data Mart and only UNION
(Multiproviders).

LSA++

Reporting on Propagator is allowed.

Consequently, necessary transformation are moved from the Business Transformation
Layer into Query Design and are thus executed upon query execution.
Virtualization layer on top of both Architected Data Marts and Propagator and using
UNION (Multiproviders) and JOIN (Composite Provider).

Streamline the Consistent EDW

Reducing number of persistent provider, esp. InfoCubes.

Optimized design and implementation of persistent providers.

Reducing change impact on persistent providers.

Streamlined EDW: Virtualization of

InfoCubes
Architected
Data Mart
Layer

Obsolete: InfoCubes as Accelerator on Business Transformation Layer DSOs

US

EU

1:
1

Business
Transformati
on Layer
EDW
Propagation
Layer

AP

US

US

AP

EU

Sales

AP

US

EU

US

AP

EU

Delivery

LSA & BW on RDBMS

US

AP

EU

Sales

AP

EU

US

AP

EU

Delivery

LSA++ & BW on HANA

Source: SAP

Streamlined EDW: Query on granular data

US AP

EDW
Propagation
Layer

EU

1:
1

Business
Transformati
on Layer

Architected
Data Mart
Layer

EDW Propagation Layer as Query Target

US

AP

EU

LSA & BW on RDBMS

US

AP

EU

LSA++ & BW on HANA

Source: SAP

Streamlined EDW - Virtual Data Mart Layer

Virtualization or Persistent Join? What are your options -

Scenario: Multiple Data Store Objects need to be Joined into a single Infoproviders.

UNION in Multiproviders doesn't correspond to reporting requirements.

LSA: BW transformation and updating(overwrite of DSO)

LSA++: Composite Provider

Reporting

Reporting
Composite
Provider

Layers and Data Persistency

Reporting on Cubes vs. DSO

LSA++ Open ODS Layer: Services

BW EDW Services

Open ODS Layer as source for persistent EDW providers

Open ODS Layer Provider as virtual part of the EDW

BW Operational Data Services

Real time replication into BW - SLT
Immediate querying on any delivered data no staging into EDW necessary
(Operational BI)
Data Modeling

BW Integration Services
Transfer/Consume HANA Modeler schemas in BW and vice versa

Decision Criteria on Using Composite Provider

Granularity / Cardinality

low

Goal: Virtualization of Architectured Data Mart Layer

Replace Data Mart

Investigate

Investigate

High/same

no

Keep Data Mart

low

Transformation / Join logic

Complex

LSA++ Holistic Framework Virtual Data Marts: Virtual Wrapping &

Combining Data

Virtual Data Marts increase overall LSA++ framework flexibility and

manageability.

Source: SAP

New Artifacts with BW 7.40

Data Provisioning

Operational Data Provisioning through ODQ

Data Modeling

Open ODS View

Quick Prototyping and Direct meta-data access

Toggling between Virtualization and Persistence through Direct access
PSA elimination
Field-based DSO
Composite Provider
Automatic Generation of HANA Models

Data Federation

Smart Data Access

Operational Data Provisioning (ODP)

Enables extract once deploy

many with EOIO Quality of
Service

Time stamp based recovery

mechanism with configurable

data retention periods

Highly efficient compression

(up to 90%)

Source: SAP

ODP HANA Context

Complementary to DB Connect

Direct loading of HANA views via DTP into BW infoproviders (PSA optional)
non BW schema

ODP / ODQ

DTP

BW Data
Source

BW
analytic view
calculation view
attribute view

ODP Extractor Context

Direct loading into BW infoproviders (PSA optional)

Scheduled or real-time daemon (Auto change notification)

Synchronous RFC instead of ALE/IDOC

Flexible Recovery

Multiple Subscribers

NW
BW
field-based
DSO

DTP

BW DataSource
DTP

PS
A

PSA
Service
InfoPackage

ODP
ODQDataSource / Extractors
NW ABAP
WebAS

Keeping rolling data window for reconciliation/recovery

ODQ Monitor (t/c ODQMON vs. RSA7)

Open ODS View

Virtual Access

Persistent Staging

Query

Query

Open ODS Layer

Open ODS Layer

Open ODS View

Open ODS View

Field based DSO

BW on
HANA

BW on
HANA
DataSource
DB/Virtual Table

ODP DataSource
DTP w/o PSA

SAPI, DB Connect, SDA

Operational Delta Queue

(ODQ), SLT, SAPI

Table / View
Table

Table

Source: SAP

ERP / BW / HANA / anyDB* Source System

Open ODS View Summary & Options

BW 74 - SP 8 / SP9

Source: SAP

Advance DSO

Source: SAP

Composite Provider

Single BW artifact to combine data from other

HANA based applications on the same instance
of HANA (consumption of native HANA models
or tables) with data in BW

Faster Implementations
More flexibility Unions and Joins
Easier consumption of data
HANA optimized query runtimes

Modern Eclipse based UI

Option to include Inventory key figures

Possibility to include in planning scenarios

Source: SAP

Automatic HANA View Generation

Info cube

Info objects

DSO

Query

Composite Provider

Use Case: Scan/POS Data Integration

Quofore / DWH
Customer Master Data
P Store No - Account.StoreChainNo
P Banner Account.Primarygroup_ID
Store Name Account.AccountName
Territory ID - Territory.Territory_ID
Territory Name
Region Customer Level 1
Zone Customer Level 2
State Address.State_ID
P Sector Name Sector Name
Product Master Data
P EAN Unit EAN Code
Category Product.ProductCategory_ID
Pack Group Product.BrandSub_ID
Product Name -Product Name
SKU SKUCode
Sector Name Sector Name
Activity Transaction Data
P Activity Name Survey.SurveyName
P Store No - Account.StoreChainNo
Display Type - ???????????
Location - ????????????

P = Primary Key

Scan Data Reporting

Data Sources Relationship
Diagram

Nielsen
Store Scan Data
P Store No - STORECODE
P EAN ITEMCODE
Sales Units SALESUNITS
Price - PRICE
Week End Date MBDCODE (First 6 Digits)

Activity Master Data Spreadsheet

Activity Master Spreadsheet Header Data
P Activity Name
Banner
Start Date
End Date

Activity Master Spreadsheet Activity Group

Activity Sub-Group Name
Activity Sub-Group Code
EAN 1-40
Activity Master Spreadsheet Detail
Activity Sub-Group Code
Store No
Target W1-12

Conclusion

SAP BW 7.40 SP8+ provides many new artifacts to:

Quickly consume external data in a direct or lightweight
persistent manner.
Combine and integrate such external data with existing BW
models using union and join operations that are HANAoptimized
Provide a new robust provider/subscriber framework for
data provisioning with reduced latency and persistency or
direct access

THANK YOU FOR PARTICIPATING

Please provide feedback on this session by completing
a short survey via the event mobile application.
SESSION CODE: BI2241
For ongoing education on this area of focus,
visit www.ASUG.com