Information Technology Reference
In-Depth Information
represented by black rectangles. With mediation transitions, a is
decomposed to a 1 and a 2 , and b is decomposed to b 1 and b 2 .
3.2.3 Choice Composition Rule
This rule is related to data generalization as explained via Figure 3.6. In
a business domain, we have a requirement A
!
B, while in a service
domain, we have two operations
f
a 1 !
b
;
a 2 !
b
g
( a
¼
a 1
[
a 2 ,
a 1
\
a 2
¼ 1 ). Then, requirement A
!
B can be realized by the XOR
(exclusive OR) collection of
, with two additional
mediation transitions as represented by black rectangles. With media-
tion transitions, data type a is specified to either a 1 or a 2 .
As shown in the composition rules we illustrate in this section,
there are direct mappings between business data and service data.
Nevertheless, usually there is no existing service data that realize
some business data. Hence, we need to add newly created service
data to make the data structure coherent. These newly created
service data can be regarded as a virtual data type to facilitate service
composition. For example, in Figure 3.5, if there is no service data type
a ( b ) that directly realizes A ( B ), we should create virtual data type a and
b from the partial realization data types of A and B , s.t. a
f
a 1 !
b
;
a 2 !
b
g
¼
a 1
a 2 ,
b
¼
b 1
b 2 , and then refine A
!
B to the AND collection of
f
b 2 g . Another circumstance is that, if a , b , a 1 , and b 1
exist, but a 2 and b 2 do not exist in the service portfolio, we should
manually add a 2 and b 2 as virtual data types in the service domain to
indicate that there is a missing operation a 2
a 1 !
b 1 ;
a 2 !
!
b 2 in the service portfolio
to fulfill the requirement a
!
b . A virtual data type can be a powerful
Data relation
Requirement
A
B
B
A
Business
domain
Service
domain
Refined
process
a
a
1
1
b
a
a
2
a
b
a
2
Service portfolio
Figure 3.6
Choice composition rule.
Search WWH ::




Custom Search