Information Technology Reference
In-Depth Information
services. Thus, the algorithm is tractable given the present environment
we have. In addition, the algorithm can generate all the feasible
composition solutions. If one needs to produce one solution only, a
polynomial-time algorithm can be easily obtained by some minor
modifications of the algorithm, since the number of available services
is finite (e.g.,
.
There have been some polynomial-time methods for automatic service
composition [153,155]. They can be employed here because our method
is a higher-level approach that does not specify a concrete reasoning
method for services in the process of composition. As a matter of fact,
we pay more attention to an idea to take the key factors as user/domain
preferences into service composition. The introduction of branch
structures is based on user/domain preferences in the requests. In the
reasoning process, once the concerned services appear, a branch
structure reflecting the user preference will be introduced after the
truth value of a concerned condition is acquired. For a specific request,
if there exist composite solutions that contain branch structures to
reflect the user preference based on the available services, they can be
constructed by this method. In addition, the introduction of branch
structures does not increase the overall complexity of the approach.
jCj
), and the length of composite services is limited by
jCj
6.4 WEB SERVICE COMPOSITIONWITH
PARALLEL STRUCTURES
In the process of service composition, if some component services
participating in a composite service can be executed in parallel, we
should seek the maximal parallelism in their execution. This can improve
the efficiency of composite service execution in a distributed and parallel
computing environment such as cloud computing. Conventionally, these
component services that can be executed in parallel are arranged in a
sequence in a feasible solution. As an undesired side effect, this leads to a
number of feasible solutions that include the same collection of compo-
nent services. The only difference among them is the execution order of
these services. Since they are actually the same, such methods likely
increase the solution cost and complexity, and contribute to low efficiency
of the composite service. Take the “treating a stroke patient” composite
service in Figure 6.2 for example. By existingmethods such as References
[152-154], 4!
¼
24 feasible solutions will be derived. They are virtually
