Top Banner
Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto University of Toronto Turning Data into Turning Data into Information Information
28

Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

Mar 29, 2015

Download

Documents

Jerome Burden
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

Dr. Eva Swenson, University of Toronto

Terry Riopelle,Software AG

Nov. 2001

University of TorontoUniversity of TorontoTurning Data into InformationTurning Data into Information

Page 2: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

2

The Business

Higher Education 5th largest university in North America Multiple IS groups reporting to multiple VPs Multiple heterogeneous platforms (OS/390, Windows, UNIX)

BUSINESS NEED – 1 Automate cross-functional processes through information sharing

across functional systems

BUSINESS NEED – 2 Deliver services directly to 54,000 students on multiple channels while

allowing 1,000 administrative staff to perform registrarial functions

Page 3: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

3

Background - The Applications

SAP- Financial System SAP- Human Resources Research System Development System SAP- Sales & Distribution Work Order other legacy systems ..

Student biographic info Admissions Registration Fees assessment Student accounts receivable Awards Academic audit Transcripts Convocation

AMS ROSI

Page 4: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

4

Background - The Technology

ROSI

MVS on mainframe

DB2

Software AGapplication

3270 terminal(non client server)

Natural

AMS

AIX on IBM SP2

ORACLE

SAP applications

SAP(client server)

WEB

ABAP

Page 5: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

5

Business Need - 1

Cross-Functional Processes

Page 6: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

6

Business Need - 1 (Cross-Functional Processes)Initial Cross-Functional Process

Institution needs to know all about an individual

Functional Systems involved SAP/HR: all about employees ROSI: all about students

Challenges Maintain data consistency between functional systems Enable SAP/HR to ask ROSI to validate student status and enable

ROSI to ask SAP/HR to validate employee status

Technical Environment Natural / DB2 on OS/390 with SAP R/3 on Unix

Page 7: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

7

Linking the Data

ROSIAMS

• Personnel number• Student number

Common Data

Page 8: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

8

Business Need - 2

Multiple Delivery Channels

Page 9: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

9

Business Need - 2 (Multiple Delivery Channels)

Channels involved Administrative staff access ROSI through 3270 terminal emulation

Students access their personal record on ROSI through the Web and the telephone (IVR)

Functional Systems involved ROSI: Student data

Challenge Achieve efficiency by using only one set of ROSI interfaces to feed all three access

channels

Technical Environment Natural / DB2 on OS/390 with Apache on AIX Unix

Page 10: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

10

Multiple Channels

Student

3270 Screens

Staff

Web Telephone

Page 11: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

11

The Futures Provide even more information that is focused on individuals

regardless of the data sources.

Continue to adapt the ROSI application to meet new vehicles/channels of delivery of information.

Page 12: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

12

The

Integration

Solutions

Page 13: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

Three Forms of Integration

Orchestrator (Integration Broker)

Communication Broker (Message Oriented Middleware)

Data Consistency

Multi-step Process

Composite Application

E.g. SAP HR/ROSI Integration E.g. ROSI Student Self Service

Message Oriented Middleware (MOM)

Integration

Broker

Business Need - 1 Cross-Functional Processes

Business Need - 2 Multiple Delivery Channels

Page 14: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

14

Integration Brokers vs. Message Oriented Middleware

Message Oriented Middleware

Programming Non-programming Application integration Solution

Synch & Asynch Desktop integration Platform integration Application integ. Platform dependencies

Integration Broker Data Management

Programming Total integration solution Synch & Asynch Information integration Platform independent Application independant

Business Need - 1 Cross-Functional Processes

Business Need - 2 Multiple Delivery Channels

Page 15: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

15

The

Business Need - 1

Cross-Functional Processes

Solution

Page 16: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

16

ROSI & SAP R/3 HR Enterprise Application Integration

Page 17: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

17

ROSI - SAP HR Integration

Page 18: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

18

What is involved?

The Process

Business objects are defined as messages

Messages are associated with an application

Integration is described as integration flows

Integration flows are associated with application integration

by defining processing, routing and delivery rules for messages

Page 19: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

19

Scenario 1

Integration Problem: Currently when an end user of the ROSI student information adds or modifies

employee number information in the ROSI financial information screen, no validation of the employee number occurs. ROSI end users would like the ability to validate the employee number given.

Solution: Allow end users from within ROSI the means to validate that the employee

number provided exists and that the student is a current employee. This will be achieved by propagating the SAP HR desired data asynchronously to a ROSI DB2 table whenever new employee information is added, or whenever an existing employee’s information is updated in the SAP HR system, which affects the applicable course fee discounts.

Page 20: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

20

Scenario 1 Integration Flow

Page 21: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

21

Scenario 2

Integration Issue: The scope of this business process is to automate this validation process such that

when a student number is added or modified in the SAP HR system. Currently when a student number added no validation of the student number occurs. HR IS end users would like the ability to validate the student number given.

Solution: To automate this validation process such that when a student number is added or

modified in the SAP HR system, EntireX Orchestrator will asynchronously validate the student number in ROSI. Validation results will be stored for auditing by HR personnel. For this validation process to occur, some transformation and filtering of data will be necessary.

Page 22: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

22

Scenario 2 Integration Flow

Page 23: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

23

The

Business Need - 2

Multiple Delivery Channels

Solution

Page 24: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

24

Multiple Channels Solution

Student

3270 Screens

Staff

Web Telephone

Page 25: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

25

What is involved?

The Process

Type of communication model chosen (RPC, ACI, event, polling, etc…)

Interfaces requirements defined (inputs, outputs)

Logical Services created

Servers created which perform the various work requested by clients for the defined Services

Clients are created in the desired languages and ask the various Services to do work.

Page 26: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

26

Multiple Channels Solution - Web

I

Physical

DB/2

Natural

EntireXBroker

Data Services Tier

NaturalNatural

Program

OS/390Mainframe

Logical Services

Logical

Presentation/Business Services Tier

AIX ApacheWeb Server

TCPIP

WebBrowser

Apache Web Server

usingCGI

Page 27: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

27

Multiple Channels Solution - IVR

I

DB/2

Natural

EntireXBroker

Data Services Tier

Physical

NaturalNatural

Program

OS/390Mainframe

Logical Services

Logical

Presentation/Business Services Tier

AIX

TCPIP

IVR

Periphonics

Page 28: Dr. Eva Swenson, University of Toronto Terry Riopelle, Software AG Nov. 2001 University of Toronto Turning Data into Information.

28

Questions?