DataXtend Semantic Integrator A Babel Fish for Your ESB

1
DataXtend Semantic IntegratorA Babel Fish for
Your ESB
Doug Merrett
Senior Solution Engineer EMEA Progress Software
2
Agenda
DataXtend Semantic Integrator A Babel Fish for
Your ESB

• Babel fish?
• What is DXSI?
• ACC Capital Holdings profile
• The business problem
• Why was DXSI chosen?
• The result
• QA

3
What the heck is a Babel fish?
The Babel fish is small, yellow and leechlike,
and probably the oddest thing in the Universe. It
feeds on brainwave energy received not from its
own carrier but from those around it. The
practical upshot of all this is that if you stick
a Babel fish in your ear you can instantly
understand anything said to you in any form of
language.
Douglas Adams The Hitchhikers Guide to the Galaxy
4
(No Transcript)
5
(No Transcript)
6
Why would I want a Babel fish for my ESB?
Solving the point-to-point transformation
problem


Transformation

• Sonic ESB eliminates the rigidity and fragility
  of point-to-point connections
• How do I solve the similar problem in my data
  model transformations?

7
Agenda
DataXtend Semantic Integrator A Babel Fish for
Your ESB

• Babel fish?
• What is DXSI?
• ACC Capital Holdings profile
• The business problem
• Why was DXSI chosen?
• The result
• QA

8
Why Semantic Integration?
How do I operationalize industry standard data
models to drive system integration?
How do I make the data models in my environment
as loosely coupled as SOA has made my services?
How do I manage data in motion to respond to
demands for real time information?
How do I reduce the effort to integrate new
applications by 50 or more?
How do I manage change to the data models used in
my services and data sources?

• How do I onboard new vendor partners in days
  instead of weeks?

9
You need a Babel fish!
Common model management



Transformation

• Rapidly create common data model from any number
  of schema
• Design mappings in Eclipse-based Workbench
• Common model mediates transformations
• Deployed and managed as an ESB service
• Java-based validation and transformation
• Regenerated if schema or common model change

10
DataXtend Semantic Integrator
Product Capabilities
DISTRIBUTEDRUNTIME
CENTRALIZEDDESIGN
TESTING
Deploy
Generate

• Visual tool for designing integration

• Stateless for scalability
• Runs in any Java container
• Connects to any bus

Runs in Eclipse
11
DataXtend Semantic Integrator
Centralized design
Map Once to Common Model
12
DataXtend Semantic Integrator
Centralized design
Visualize Large, Complex Models
13
DataXtend Semantic Integrator
Centralized design
Create Visual Mappings to and from the common
model
14
DataXtend Semantic Integrator
Centralized design
Virtual attributes simplify models without
deviating from the standard
15
DataXtend Semantic Integrator
Centralized design
Multiple addresses required for a service move
16
DataXtend Semantic Integrator
Testing
17
DataXtend Semantic Integrator
Deployment and impact analysis
Impact Analysis Report Details how information
is used in addition to where it is used
18
Credit Mappings in DXSI
Integration Projects
Property Sheet
Main Panel
ModelOverview
Editor, Impact Analysis, Tester
19
MISMO - OAGIS Credit Response Class Maps
20
Agenda
DataXtend Semantic Integrator A Babel Fish for
Your ESB

• Babel fish?
• What is DXSI?
• ACC Capital Holdings profile
• The business problem
• Why was DXSI chosen?
• The result
• QA

21
Ameriquest Mortgage Company
Mortgage origination system SOA infrastructure

• Situation
• One of the largest mortgage companies, originates
  and services home mortgage loans through more
  than 280 branch offices across the United States.
  Ameriquest and its affiliates employ more than
  17,000 associates nationwide.
• Grew from 35,600 loans in 2000 to 118,900 loans
  in 2003. Increased loan volume from 3.7 billion
  in 2000 to 17.8 billion in 2003.
• Need to rapidly rollout new mortgage products in
  response to changes in interest rates and market
  conditions.
• New loan origination system deployment required
  scalability, reliability, and flexibility beyond
  that of available with existing SOA
  infrastructure built on BEA WebLogic.

22
VMS RR Overview
Vendor Management System Request Response

• VMS RR provides
• B2B integration with business partners (Credit,
  AVM, Title, etc.) to complete loan origination
  for Ameriquest Argent Mortgage.
• Implements a SOA approach using
• Sonic ESB
• DataXtend Semantic Integrator
• OAGIS Open Application Group Integration
  Specification BOD XML business messages
• MISMO Mortgage Industry Standards Maintenance
  Organization XML standard for data exchange with
  business partners

23
VMS RR Context Diagram
Vendor Management System Request Response
24
Agenda
DataXtend Semantic Integrator A Babel Fish for
Your ESB

• Babel fish?
• What is DXSI?
• ACC Capital Holdings profile
• The business problem
• Why was DXSI chosen?
• The result
• QA

25
Problem Statement

• Unable to deliver a B2B integration with our
  business partners that was acceptable to our
  business users
• Monthly run rate was approx. 250K-500K
• Required 14 resources to deliver 4 out of 14 B2B
  integrations
• Average time to market was 4 months per
  integration

26
Success Criteria

• Project Requirements
• Reduce cost of ownership
• Improve time to market
• Business Drivers
• Reduce head count
• Improve business users and partners satisfaction
• Automate the development process and generate
  code
• Establish common and repeatable method to
  implement each integration
• Allow the business system analyst (BSA) to drive
  development process
• Validate quality early in the process (QA)

27
Agenda
DataXtend Semantic Integrator A Babel Fish for
Your ESB

• Babel fish?
• What is DXSI?
• ACC Capital Holdings profile
• The business problem
• Why was DXSI chosen?
• The result
• QA

28
Proof of Concept
VMS RR and DataXtend Marriage

• Learning curve was low, BSA became primary user
• BSA drove the development process
• Generated consistent stateless JAVA code and
  required minimum coding
• QA participated early in the process
• Passed performance test criteria
• Deployment to our environments was painless
• Integrated seamlessly with Sonic ESB

29
Selling the Organization
Addressing constituent concerns

• Business Users
• Faster implementation time
• Flexible and adaptable to user changes
• CFO
• Product ROI 1 month
• CIO
• Reduced development cost benefits
• Reduced resources required to create XML
  transformations
• Reduced time to produce XML transformations
• Improved quality benefits
• Fewer defects, resulting in less rework
• Fewer incidents, resulting in less maintenance
  support costs
• Reduced time spent handling customer complaints
• Architects Software Engineers
• More time working on interesting projects less
  time on XSLT

30
VMS RR Architecture
This is a façade for all DXSI services. This and
the product specific data services can be
deployed on the same container instance as VMS.
Product specific data service implementations. A
data service can perform validation,
transformation or other data manipulation.
31
VMS RR Credit Check Flow
4. Credit Vendor Request
1. Credit Check Order Placed
5. Credit Vendor Response
8. Credit Check Complete
Vendor Management System Request Response
2. Validate Credit Order Request
3. Map Credit Order Request
6. Validate Credit Order Response
7. Map Credit Order Response
32
Agenda
DataXtend Semantic Integrator A Babel Fish for
Your ESB

• Babel fish?
• What is DXSI?
• ACC Capital Holdings profile
• The business problem
• Why was DXSI chosen?
• The result
• QA

33
VMS RR and DataXtend Results

• Reduced from 14 to 4 resources
• 300 improvement
• Cost of ownership reduced from 250K to 20k
• 1250 improvement
• Time to market now 2-4 weeks from 14 weeks
• 400 improvement
• Implemented all 14 integrations
• Completed the 10 outstanding integrations
• Became more flexible and agile Capable of
  adapting to the business change requests
• Demonstrated a successful SOA implementation

34
In Summary
DataXtend Semantic Integrator at ACC Capital
Holdings

• Eliminates the semantic mismatch between services
  in SOA
• Reduces the time-to-market for adopting industry
  data models like OAGIS, MISMO
• Dramatically reduces the cost of service
  development and on-going maintenance of SOA
  projects
• Is vital part of keeping SOA agile for future
  change

35
Questions?
For more information Visit us at
www.progress.com/dataxtend
36
(No Transcript)