Information Technology Reference
In-Depth Information
takes advantage of vehicles with cooperative communication skills, in a way that these
cars are able to react to possible accident risks or emergence situations. The CCA mecha-
nism generates an encapsulated notification which is sent as a message through a one-hop
communication scheme to all vehicles within a potential danger coverage (relay schemes
are also possible). It should be noted that the establishment of this VANET application
will be deployed gradually, equipping vehicles with the proper hardware and software
so as they can communicate in an effective way within the vehicular environment.
In our research we consider a platoon (or chain) of
N
vehicles following a leading
one. The leading vehicle stops instantly and the following vehicles start to brake when
they are aware of the risk of collision, because of a warning message reception or the
perception of a reduction in the speed of the vehicle immediately ahead. To test the
worst case situation, vehicles cannot change lane or perform evasive maneuvers.
We have developed a first approach mathematical model to calculate the average
percentage of accidents in the platoon, varying the number of considered vehicles,
their average speed, the average inter-vehicle spacing and the penetration ratio of the
CCA technology. Specifically when the CCA penetration ratio is taken into account, the
growth in the number of operations of the analytical model is such that the sequential
computation of a numerical solution is no longer feasible. Consequently, we resort to
the use of the OpenMP parallelization techniques for solving those computational cases
considered as unapproachable by means of sequential procedures.
Additionally, we execute our programs in the Ben-Arabi Supercomputing environ-
ment [2], taking the advantage of utilizing the fourth fastest Supercomputer in Spain.
In the current work we show how the parallelization techniques coordinated with su-
percomputing resources make the simulation process a more suitable and efficient one,
allowing a thorough evaluation of the CCA application.
The remainder of this paper is organized as follows. In Section 2 we briefly review
the related work. In Section 3 the OpenMP environment is briefly reviewed and the
Ben-Arabi Supercomputer architecture introduced. A description of the mathematical
model, its implementation and parallelization are provided in Sections 4 and 5. Finally,
some results are shown and discussed in Section 6 to illustrate the performance of the
resulting parallel algorithm. In this section it is also described our unsuccessful experi-
ence of using the MPI parallelization technique to further reduce the computation times.
Conclusions and future work are remarked in Section 7. Let us mention that this paper
is an extension of the work presented by the authors in [3].
2
Related Work
So far, most typical High Performance Computing (HPC) problems focused on those
fields related with certain fundamental problems in several areas of science and engi-
neering. Other typical applications are the ones related to commerce, like databases and
data mining [4]. That is the reason why we consider our VANET mathematical model
approximation as a non-classical issue to be solved under HPC conditions, contributing
to extend the use of supercomputing to other fields of interest.
In the implementation of our mathematical model we parallelize a sparse matrix-
vector multiplication. This operation is considered as a relevant computational kernel in
Search WWH ::
Custom Search