IBM Branch Transformation Toolkit 5.1 Migration and Usage Guidelines

Running H/F 1 289

3. Locate the keyed collection definition with the ID Formats, and replace the

entire keyed collection definition with the one listed in Example 8-4.

Example 8-4 The correct formats keyed collection definition

<kColl id="formats" dynamic="false" >

<field id="fTypedData" value="com.ibm.btt.base.TypedDataElementFormat" description="" />

<field id="codeSetTrans" value="com.ibm.btt.base.CodeSetTranslator" description="" />

<field id="compress" value="com.ibm.btt.base.Compressor" description="" />

<field id="constant" value="com.ibm.btt.base.ConstantFormat" description="" />

<field id="delim" value="com.ibm.btt.base.Delimiter" description="" />

<field id="fDate" value="com.ibm.btt.base.DateFormat" description="" />

<field id="fFloat" value="com.ibm.btt.base.FloatFormat" description="" />

<field id="fInteger" value="com.ibm.btt.base.IntegerFormat" description="" />

<field id="fixedLength" value="com.ibm.btt.base.FixedLength" description="" />

<field id="fNumString" value="com.ibm.btt.base.NumericStringFormat" description="" />

<field id="fObject" value="com.ibm.btt.base.ObjectFormat" description="" />

<field id="fString" value="com.ibm.btt.base.StringFormat" description="" />

<field id="fTime" value="com.ibm.btt.base.TimeFormat" description="" />

<field id="fXML" value="com.ibm.btt.base.XMLFormat" description="compound" />

<field id="hashtable" value="com.ibm.btt.base.HashtableFormat" description="compound" />

<field id="hashtableIColl" value="com.ibm.btt.base.HashtableIndexedCollectionFormat"

description="compound" />

<field id="iCollF" value="com.ibm.btt.base.IndexedCollectionFormat" description="compound"

/>

<field id="id" value="com.ibm.btt.base.Identifier" description="" />

<field id="map" value="com.ibm.dse.services.ldap.MapFormat" description="compound" />

<field id="mapper" value="com.ibm.btt.base.DataMapperFormat" description="compound" />

<field id="mapperConverter" value="com.ibm.btt.base.DataMapperConverterFormat"

description="" />

<field id="mapping" value="com.ibm.dse.services.ldap.Mapping" description="" />

<field id="maxLength" value="com.ibm.btt.base.MaximumLength" description="" />

<field id="nullCheck" value="com.ibm.btt.base.NullCheckDecorator" description="" />

<field id="record" value="com.ibm.btt.base.RecordFormat" description="compound" />

<field id="refFmt" value="java.lang.Object" description="" />

</kColl >

4. Press Ctrl+S to save the change.

8.3.5 The Branch Transformation Toolkit architecture

The IBM IBM Branch Transformation Toolkit for WebSphere Studio is a

component-based toolkit for developing enterprise e-business applications. The

Branch Transformation Toolkit enables the development of interfaces to the

services of a financial institution's information system so that they become

ubiquitous through all delivery channels such as the traditional branch, call

center, banking kiosk, Internet banking, and mobile access. This minimizes the

290 IBM Branch Transformation Toolkit 5.1 Migration and Usage Guidelines

necessity to develop new code and reduces the time required to make new

financial services available to all delivery channels.

The architecture and technological approach of the Branch Transformation

Toolkit creates retail delivery solutions that preserve investment in existing

enterprise systems while accounting for the inherent instability of any

infrastructure due to the innovations that appear frequently in the high-tech

industry. While providing a way to preserve the existing systems, the Branch

Transformation Toolkit is not tied to one particular platform since it is built on

Java

TM

, the programming language of choice for handling platform change.

The toolkit also takes advantage of existing platforms and technologies such as

Eclipse, Web services, J2EE, and so on. The toolkit runtime architecture is

based on the J2EE architecture with extensions, and many development tools

the toolkit provides are Eclipse plug-ins.

Architecture

The architecture of the toolkit’s application is based on a logical three-tier model,

that is, back-end enterprise tier, application server tier, and client tier.

Within the application server tier, the toolkit has two separate layers. The

application presentation layer is responsible for receiving requests from the client

and passing that request to the application logic layer. It also passes the

response back to the client. The application logic layer is responsible for

performing the request as a process and passing the response back to the

application presentation layer.

In general, the application presentation layer resides in a Web container in

WebSphere Application Server, while the application logic layer resides in a EJB

container. The services are the exception because they can reside anywhere.

The design and portability of the toolkit allow the middle-tier servers to exist at

either the branch level, that is, one server per branch, the regional level, that is,

one server per group of branches, or even a centralized level, that is, a single

server for the entire financial institution. The design provides flexibility to achieve

the right balance between the number of servers and the network bandwidth,

without affecting any application logic. Besides the application server, there

might be a

technical server responsible for providing common services such as

disks or printers to the client workstations. If the application presentation layer

and the application logic layer are on the same architectural level, they can

physically be on the same machine.

See Figure 8-44 on page 291.

Running H/F 1 291

Figure 8-44 Three-tier architecture

Client

A client in the three-tier architecture contains little logic. The logic it does have is

usually presentation logic or logic required locally to do such things as access

Client tier

BTT data

context

Java client

BTT panel (AWT)

Application server tier

Application presentation layer

(Web container)

EJB

Application logic layer

(EJB container)

EJB

Struts action

Struts action servlet &

request processor

View

Struts form

BTT data

context

EJB

WSIF/

EJB

BTT data

context

BTT EJB

Process Choreographer

Micro flow

BTT

BTT data

context

BTT client side

micro flow

Services

WSIF

Invokers

WSIF/

EJB

Back-end enterprise tier

JDBC

database services

JCA resource

adapters

Data

WSIF