Real-Time Data Warehousing

Published on July 2016 | Categories: Documents | Downloads: 59 | Comments: 0 | Views: 648
of 39
Download PDF   Embed   Report

Real-Time Data Warehousing with SAP NetWeaver Business Intelligence

Comments

Content

HTWG: AK-SWT Treffen 10.11.2006

Real-Time Data Warehousing with SAP NetWeaver Business Intelligence
Rainer Uhle, SAP NW BI Produktmanagement

Systemlandschaftsbetrachtung zum Thema Integration

MDM
(M/R-DBMS)

BI (OLAP)

Analyse von Prozessen • FI/CO
• Einkauf, Fertigung, Vertrieb • HR

Ziel

Stamm- und Prozessdaten Replikation • periodisch
• ETL • Real-time Data Acquisition •… • stetig • EAI •… • sonstig • Direct Access •…

• Entscheidungsunterstützung (DS)
• Planung

SAP Terminologie

ERM
(RDBMS)

ERP
(OLTP)

Abbildung von Prozessen • FI/CO
• Einkauf, Fertigung, Vertrieb • HR

 SAP AG 2006, HTWG AK SWT 101106 / 2

RAS Core Research Note G00130597, Bill Gassman, 30 September 2006

Options Proliferate for ‘Real-Time Data Integration’ Technology • Numerous applications are emerging for real-time or low-latency data integration. Consequently, IT development and operations managers must identify the most cost-effective and appropriate offerings to satisfy on-demand and event-driven needs. • Analysis: Real-time data integration is the technology and process that provides application access to up-to-date information to meet time-sensitive business requirements. • Technology Options • direct access • real-time data acquisition … • publish-subsribe model, change data capture, message exchange • master data centralization • File transfer protocol (FTP) •…

 SAP AG 2006, HTWG AK SWT 101106 / 3

Challenges of Real-time Data Warehousing
 Looking beyond traditional ETL expand the concept of data acquisition in order to reduce data latency  Latency – time lag between an activity completion in an ERP environment and the availability of the completed activity data in a state of the art data warehouse environment  Latency = zero demand for tools that enable direct access of data and information without any latency across platforms  SAP NetWeaver 2004s BI capabilities for Real-time Data Warehousing
– Real-time Data Acquisition (RDA) – Direct Access

 SAP AG 2006, HTWG AK SWT 101106 / 4

ETL, Flavours of ETL, Alternatives ETL (Extraction, Transformation, Load)
 New DataSource Concept with SAP NW2004s BI  DataSource  Transformation  Data Transfer Process (DTP)

Flavours of ETL
 Classical Staging Processes  Real-time Data Acquisition (RDA)

Alternatives
 Direct Access
How does the Implementation of ETL Processes look like in an SAP NetWeaver 2004s BI Environment?

 SAP AG 2006, HTWG AK SWT 101106 / 5

Real-time Data Warehousing - Introduction Overview NW 2004s BI - ETL Capabilities Real-time Data Warehousing with SAP NW 04s BI Direct Access Real-time Data Acquisition (RDA) Outlook and Summary

Real-time Data Warehousing - Introduction Overview NW 2004s BI - ETL Capabilities Real-time Data Warehousing with SAP NW 04s BI Direct Access Real-time Data Acquisition (RDA) Outlook and Summary

ETL, Flavours of ETL, Alternatives ETL (Extraction, Transformation, Load)
 New DataSource Concept with SAP NW2004s BI  DataSource  Transformation  Data Transfer Process (DTP)

Flavors of ETL
 Classical Staging Processes
– Persistency of transactional and master data in BI – Extraction using DataSources with services like packaging and sequencing, delta handling – Transformations with graphical UI and sophisticated formula builder – Data loads based on batch scheduling – Data loads can even be scheduled hourly – Process Chains for modeling complex load processes using InfoPackages and DTPs – Scheduling options for Process Chains using SAP NetWeaver batch scheduling or Redwood Chronacle (OEM)

 SAP AG 2006, HTWG AK SWT 101106 / 8

SAP NW2004s BI: Data Flow Concept – Example

SAP NetWeaver BI
InfoProvider

Process Chain

Data Transfer Process

TRANSFORMATION

DataSource (PSA) InfoPackage

Source System
 SAP AG 2006, HTWG AK SWT 101106 / 9

Data Acquisition Layer – Data Source/Source Systems

Create DB Connection and DataSource

DataSource
DB Connect UD Connect BI Service API File Interface Web Service

Relational Source

MultiDimensional Source
e.g. Hyperion

SAP Source

File

XML

e.g. MSS, DB2, Teradata

e.g. SAP ERP

 SAP AG 2006, HTWG AK SWT 101106 / 10

New BI DataSource concept with NetWeaver 2004s Highlights
 unique look and feel for all of the DataSource Types  direct/remote access is always an option

 preview feature is standard
 automated coversions (e.g. date format detection)  InfoPackages only write into PSA  new Transformation handling

 new Data Transfer Process logic

 SAP AG 2006, HTWG AK SWT 101106 / 11

Source System Tree
Source Sytems categorized:
 SAP vs. non SAP  File vs. database  Relational vs. Multidimensional DB  ABAP vs. Java  XML vs. Text/Binary

 SAP AG 2006, HTWG AK SWT 101106 / 12

SAP NW2004s BI: Data Flow Concept – Example

SAP NetWeaver BI
InfoProvider

Process Chain

Data Transfer Process

TRANSFORMATION

DataSource (PSA) InfoPackage

Source System 1
 SAP AG 2006, HTWG AK SWT 101106 / 13

Transformation – Graphical UI

Source fields

Target fields

Note: Key figures, characteristics and date fields are shown on the same level (transformation group)
 SAP AG 2006, HTWG AK SWT 101106 / 14

Rules per group

Transformation - Rules

 Information on
– Rule type – Currency/ Unit Conversion – Source fields – Target fields

 SAP AG 2006, HTWG AK SWT 101106 / 15

Data Transfer Process: Complex Example SAP Netweaver BI
InfoProvider

Process Chain
Process Chain
DTP

TR
DataStore Object 3

DTP

DTP

TR DataStore Object 1

TR

DTP

DataStore Object 2 DTP

DTP

TR DataSource (PSA)

TR
DataSource (PSA)

IP

IP

Source System 1
 SAP AG 2006, HTWG AK SWT 101106 / 16

Source System 2

Complex Implementation Example

 SAP AG 2006, HTWG AK SWT 101106 / 17

Real-time Data Warehousing - Introduction Overview NW 2004s BI - ETL Capabilities Real-time Data Warehousing with SAP NW 04s BI Direct Access Real-time Data Acquisition (RDA) Outlook and Summary

Two given Options for real-time Data Warehousing
ETL (Extraction, Transformation, Load)
 New DataSource Concept with SAP NW2004s BI  DataSource  Transformation  Data Transfer Process (DTP)

Flavours of ETL
 Classical Staging Processes  Realtime Data Acquisition (RDA)
– Integrated in data flow concept – DataSources with specific Adapter Types like ‚WebService Push‘ and ‚Real-time Extraction from SAP System‘ – DTP Type ‘Real-time Data Acquisition‘ – Daemon based processing with modified Request Handling

Alternatives
 Direct Access
– – – – Modeling based on VirtualProvider DataSources are Direct Access enabled DTP Type ‘for Direct Access‘ Driven by Query handling

 SAP AG 2006, HTWG AK SWT 101106 / 19

Real-time Data Warehousing - Introduction Overview NW 2004s BI - ETL Capabilities Real-time Data Warehousing with SAP NW 04s BI Direct Access Real-time Data Acquisition (RDA) Outlook and Summary

Direct Access: Simple Example
Query

BI
VirtualProvider

DataTransfer Process for Direct Access

TRANSFORMATION

DataSource for direct access

SourceSystem

 SAP AG 2006, HTWG AK SWT 101106 / 21

Direct Access - Implementation

Create VirtualProvider

Create Transformation and DTP

 SAP AG 2006, HTWG AK SWT 101106 / 22

Direct Access: Complex Example
Query

BI

VirtualProvider

Remote Characteristic as InfoProvider

DTP for Direct Access

TR

TR

DTP for Direct Access

DataSource (direct access)

DataSource (direct access)

Source System

Transactional Data

Master Data

 SAP AG 2006, HTWG AK SWT 101106 / 23

Implementation - Overview

Verify master data handling during Reporting

Create Remote Characteristic as InfoProvider
 SAP AG 2006, HTWG AK SWT 101106 / 24

Final Query Result with Remote Master Data Texts

Customer Remote names from Customer external source Names

 SAP AG 2006, HTWG AK SWT 101106 / 25

Direct Access Scenarios
BI DataSource Types with Direct Access capabilities File DataSource
 Access file system of application server or local workstation

DB Connect DataSource
 Access SAP NetWeaver DB platforms using ODBC technique

UD Connect DataSource
 Based on the Universal Data Integration (UDI) concept  Access to any source system using SAP J2EE Connection Framework based on BI Java Connectors
– JDBC Connector for any DB providing JDBC (MS Access, Teradata, …) – ODBO Connector for any multi dimensional providing OLE DB for OLAP (Hyperion, …) – XML/A Connector for sources providing XML for Analysis (SAP NW BI, …)

BI Service API DataSources
 Access to SAP ERP source systems

 SAP AG 2006, HTWG AK SWT 101106 / 26

Real-time Data Warehousing - Introduction Overview NW 2004s BI - ETL Capabilities Real-time Data Warehousing with SAP NW 04s BI Direct Access Real-time Data Acquisition (RDA) Outlook and Summary

Strategic vs. Tactical Decision-Making

Standard Data Acquisition Goal Strategic decisionmaking (long term planning) Request oriented (nightly batch job) 1/day … 1/week After a certain period

Real-time Data Acquisition Tactical decision-making (day-to-day decisions) Stream oriented (always active batch job) 1/min … 1/hour Close to real-time

Data Staging Upload frequency Availability for reporting Computing Power

Usually done at night (load balancing)

Permanent resource consumption

 Use Real-time Data Acquisition only if necessary.

 SAP AG 2006, HTWG AK SWT 101106 / 28

Example: RDA using SAP ERP Service API

SAP NetWeaver BI

Operational Data Store
DataStore Objects Daemon Pull ~ 1/min
Data Transfer Process for Real-time Data Acquisition

Service API-based real-time data acquisition is a two-stage process:
 Data is pulled into PSA by Realtime InfoPackage  Data is transferred to DataStore Object using Real-time DTP

PSA DataSource
InfoPackage for Real-time Data Acquisition

In both cases the process of data movement is initiated by the system daemon

Delta Queue Service API

Real-time Update

Application

SAP Source System
 SAP AG 2006, HTWG AK SWT 101106 / 29

Real-time Data Acquisition (RDA) - Processing Daemon
 Originally stands for Disk And Execution MONitor in the UNIX world  System process to initiate data loads at regular intervals: from one minute to hourly  The BI Daemon data load includes three steps:
– Initiate BI Service-API data pull using InfoPackage for RDA into PSA (SAP source systems) – Track status of data transfer from source system

– Initiate update of DataStore Object using DTP

 Successful execution of each step is tracked in a control table
– Allows restarting – Restart can be initiated so it starts at the next step after the last successfully executed step

 SAP AG 2006, HTWG AK SWT 101106 / 30

Daemon Monitoring

How to check the status of RDA daemon?
 Call transaction RSRDA or press
– Is the daemon (batch job) still running?

Refresh in RSRDA

 The status of the daemon is indicated by one of the following icons:
– Daemon active and running – Daemon not active – Daemon stopped with an error – Daemon stopped by user (will change to “not active” or “erroneous”)

 Display runtime information about the daemon (context menu in RSRDA)
 SAP AG 2006, HTWG AK SWT 101106 / 31

RDA using SAP ERP Service API

SAP NetWeaver BI
DS-Object

Daemon Process

RDA-Data Transfer Process

TRANSFORMATION

DataSource (PSA) RDA-InfoPackage

SAP ECC
 SAP AG 2006, HTWG AK SWT 101106 / 32

RDA using SAP ERP Service API - Modelling View

Create InfoPackage and DTP for RDA

 SAP AG 2006, HTWG AK SWT 101106 / 33

RDA Example Using Service API – Administration View

Create a Daemon in the RDA Monitor generating an open request waiting for transactional SAP ERP PO data
Monitor: Real-time Data Acquisition

Run SAP ERP Purchasing to push a data record into BI using this chanel
 SAP AG 2006, HTWG AK SWT 101106 / 34

RDA – Implementation Scenario

BI

Enhance established Data Flow with RDA capabilities
DataStore Object
DataStore Object

 Implement additional DataStore Object for operational reporting  Replace standard delta InfoPackage by RDA InfoPackage  Regular data loads can be scheduled after closing the RDA InfoPackage Request using appropriate Process Chain feature

RDA

Daemon

PSA

DataSource
RDA

 Typically data is deleted regularly from the DataStore Object supplied using RDA  Standard reporting can be enhanced by operational reporting using the report-report interface

OLTP
2LIS_02_VAITM

real-time Update

Application

Delta Queue

 SAP AG 2006, HTWG AK SWT 101106 / 35

Real-time Data Warehousing - Introduction Overview NW 2004s BI - ETL Capabilities Real-time Data Warehousing with SAP NW 04s BI Direct Access Real-time Data Acquisition (RDA) Outlook and Summary

Real-time Data Integration with SAP NW BI

MDM
(M/R-DBMS)

BI (OLAP)

Analyse von Prozessen • FI/CO
• Einkauf, Fertigung, Vertrieb • HR

Ziel

Stamm- und Prozessdaten Replikation • periodisch
• SAP NW BI ETL • SAP NW BI RDA •… • stetig • EAI (SAP NW Process Integration) •… • sonstig • SAP NW BI Direct Access •…

• Entscheidungsunterstützung (DS)
• Planung

Complete SAP SAP Terminologie Solution

ERM
(RDBMS)

ERP
(OLTP)

Abbildung von Prozessen • FI/CO
• Einkauf, Fertigung, Vertrieb • HR

 SAP AG 2006, HTWG AK SWT 101106 / 37

Real-Time Data Warehousing with SAP NetWeaver BI Keypoints to take home  Rich standard ETL capabilities

 Direct Access
– allows querying data without data replication – any source system type (e.g. JDBC)

 Real-time Data Acquisition
– reduces data latency in the operational data store to ~ 1 min – Daemon based data loading

 SAP AG 2006, HTWG AK SWT 101106 / 38

THANK YOU FOR YOUR ATTENTION !

QUESTIONS



SUGGESTIONS



DISCUSSION

 SAP AG 2006, HTWG AK SWT 101106 / 39

Sponsor Documents

Or use your account on DocShare.tips

Hide

Forgot your password?

Or register your new account on DocShare.tips

Hide

Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in

Close