Information Technology Reference
In-Depth Information
of an MD-SPL. Finally, we present our mechanisms for
deriving configured products supported by decision models in
section 5.3. At the end of this chapter, we present limitations
of FieSta in section 5.4.
5.1.1.
Coarse-grained and fine-grained variations
The base mechanisms which we have introduced until now
in this topic allow product linearchitectstocaptureandexpress
the possible variations between members of a product line
by separately creating metamodels and feature models. This
allows us to capture and express
coarse-grained variations
between products. For example, a Smart-Home system has
a coarse-grained variation, with respect to the
Automatic
Windows
feature, if either all or none of the
Windows
in the
Smart-Home products generated from the product line are
Automatic Windows
. Rephrasing it, a coarse-grained feature
may be selected for all the models conforming to a given
concept. Thus, a coarse-grained variation applies uniformly to
all the instances of a metaconcept.
We obtain coarse-grained variations between members
of our product line example by creating coarse-grained
configurations.A
coarse-grained configuration
isanassociation
between models that conform to metamodels and instances of
feature models. Thus, for instance, a first Smart-Home system
can be coarse-grained configured by creating a domain model
andselectingthefeature
Fingerprint
.AsecondSmart-Home
system can be coarse-grained configured by using the same
domain model and selecting the feature
Keypad
. When Smart-
Home systems are derived,a coarse-grained variation between
them appears: all the
Doors
in the first Smart-Home system
have
Fingerprint
as lock door control mechanism, and all
the
Doors
in the second Smart-Home system have
Keypad
as
lock door control mechanism. One immediate solution to this
problem is to refine the model and to introduce more specific
models making some of the variations explicit. If the degree of
Search WWH ::
Custom Search