|
- LBM Training Course
- Over the last years, the Lattice-Boltzmann
Method has developed as a competitive technique
in modeling fluid flows. Compared to "classical"
methods, the LBM exhibit various advantages,
in particular when simulating flows in complex
geometries or dynamic problems. The present short
course intents to provide an overview over the current
state of the LBM technique with a focus on high
performance computing and parallelization aspects.
Novices in CFD will find useful practical aspects of
LBM. More advanced practitioners will find helpful
hints regarding optimization and tuning of their own
codes. The course also comprises hands on tutorials.
The course instructors are:
- Dr. Gerhard Wellein, RRZE Erlangen, Germany
- Dr. Thomas Zeiser, RRZE Erlangen, Germany
- Dr. Peter Lammers, Uni Stuttgart, Germany
- Prof. Manfred Krafzyck, TU Braunschweig, Germany
- Prof. Gunther Brenner, TU Clausthal, Germany
- Prof. Boris Chetverushkin, IMAMOD, Russia
Course Material:
- Parallel and Distributed Scientific Computing
- The course intends to give a survey of
modern software tools for parallel and distributed
scientific computing for the computational fluid
dynamicist. Every aspect of the short course will be
illustrated by numerical examples and software. The
course will begin with a general description of
parallel computing and survey generic tools such
as MPI. This will be followed by a presentation
of PETSc, a general purpose scientific computing
library that implements state of the art parallel
numerical algorithms and provides a framework for the
development of large scale codes. This presentation
will be given by one of the main authors of this
library, which is broadly used throughout the
world. The course will finish with a survey of
algorithms that have been developed for distributed
and grid computing, with applications to CFD.
The course instructors are:
- Dr. Matthew G. Knepley, Argonne National Laboratory, Chicago, IL.
- Hatem Ltaief, Dept. of Computer Science, University of Houston.
|