Information Technology Reference
In-Depth Information
on an event-driven algorithm, named Green's function reaction dynamics, which
uses Green's functions to combine in one step the propagation of the particles
in space with the reactions between them. In this algorithm, the particles move
diffusively; it is assumed that if a reaction exists, it follows a Poisson process and
it happens instantaneously. This means that the reaction event can be decoupled
from the diffusive motion of the particle. The time step of the algorithm is deter-
mined such that only single particles or pairs of particles have to be considered,
avoiding complex reaction rules.
MCell is used as the simulation engine; one of its outputs consists in the
position of each molecule in the system every time step. These positions are
used as the input for FERNet (Fluorescence Emission Routine Network), which
generates the fluorescent trace for each sample time. This data is then compared
against the experimental data and can support the validation of the proposed
model.
3 Volunteer Grid/Cloud Computing
This section describes the paradigms of grid and cloud computing, and the Our-
Grid middleware for developing applications over volunteer grid/cloud platforms.
3.1 Grid and Cloud Computing
Grid computing is a paradigm for parallel/distributed information processing
that allows the integration of many computer resources to provide a power-
ful computing platform that allows solving applications with high computing
demands. This paradigm has been increasingly employed to solve complex prob-
lems (i.e. e-Science, optimization, simulation, etc.) in the last twenty years [10].
Grid infrastructures are conceived as a large loosely-coupled virtual super-
computer formed by many heterogeneous platforms of different characteristics,
usually working with non-interactive workloads with a large number of files. Grid
infrastructures have made it feasible to provide pervasive and cost-effective ac-
cess to distributed computing resources for solving hard problems [5]. Starting
from small grids in the earlier 2000's, nowadays grid computing is a consolidated
field of research in Computer Science and many grid infrastructures are widely
available. As of 2012, more than 12 PFLOPS are available in the current more
powerful grid system, from the
Folding@home
project.
In the last years, cloud computing [4,8] emerged as one of the main existing
computing paradigms, mainly due to several very interesting features it provides,
including elasticity, flexibility, or large computational power, among many oth-
ers. Cloud computing has raised the interest of both academic and industrial
research communities, by providing a computing model which is able to cope
eciently with complex problems involving hard computing functions and han-
dling very large volumes of data.
Cloud computing provides a stack with different kinds of services to users,
including: Infrastructure as a Service (IaaS), dealing with resources as servers,
