On the parallel scalability of hybrid linear solvers for large 3D problems
Large-scale scientific applications and industrial simulations are nowadays fully integrated in many engineering areas. They involve the solution of large sparse linear systems. The use of large high performance computers is mandatory to solve these problems. The main topic of this research work was...
| Main Author: | |
|---|---|
| Format: | Others |
| Published: |
2008
|
| Online Access: | http://oatao.univ-toulouse.fr/7711/1/haidar.pdf |
| id |
ndltd-univ-toulouse.fr-oai-oatao.univ-toulouse.fr-7711 |
|---|---|
| record_format |
oai_dc |
| spelling |
ndltd-univ-toulouse.fr-oai-oatao.univ-toulouse.fr-77112017-10-11T05:09:25Z On the parallel scalability of hybrid linear solvers for large 3D problems Haidar, Azzam Large-scale scientific applications and industrial simulations are nowadays fully integrated in many engineering areas. They involve the solution of large sparse linear systems. The use of large high performance computers is mandatory to solve these problems. The main topic of this research work was the study of a numerical technique that had attractive features for an efficient solution of large scale linear systems on large massively parallel platforms. The goal is to develop a high performance hybrid direct/iterative approach for solving large 3D problems. We focus specifically on the associated domain decomposition techniques for the parallel solution of large linear systems. We have investigated several algebraic preconditioning techniques, discussed their numerical behaviours, their parallel implementations and scalabilities. We have compared their performances on a set of 3D grand challenge problems. 2008-06-23 PhD Thesis PeerReviewed application/pdf http://oatao.univ-toulouse.fr/7711/1/haidar.pdf info:eu-repo/semantics/doctoralThesis info:eu-repo/semantics/openAccess Haidar, Azzam. On the parallel scalability of hybrid linear solvers for large 3D problems. PhD, Institut National Polytechnique de Toulouse, 2008 http://ethesis.inp-toulouse.fr/archive/00000650/ http://oatao.univ-toulouse.fr/7711/ |
| collection |
NDLTD |
| format |
Others
|
| sources |
NDLTD |
| description |
Large-scale scientific applications and industrial simulations are nowadays fully integrated in many engineering areas. They involve the solution of large sparse linear systems. The use of large high performance computers is mandatory to solve these problems. The main topic of this research work was the study of a numerical technique that had attractive features for an efficient solution of large scale linear systems on large massively parallel platforms. The goal is to develop a high performance hybrid direct/iterative approach for solving large 3D problems. We focus specifically on the associated domain decomposition techniques for the parallel solution of large linear systems. We have investigated several algebraic preconditioning techniques, discussed their numerical behaviours, their parallel implementations and scalabilities. We have compared their performances on a set of 3D grand challenge problems. |
| author |
Haidar, Azzam |
| spellingShingle |
Haidar, Azzam On the parallel scalability of hybrid linear solvers for large 3D problems |
| author_facet |
Haidar, Azzam |
| author_sort |
Haidar, Azzam |
| title |
On the parallel scalability of hybrid linear solvers for large 3D problems |
| title_short |
On the parallel scalability of hybrid linear solvers for large 3D problems |
| title_full |
On the parallel scalability of hybrid linear solvers for large 3D problems |
| title_fullStr |
On the parallel scalability of hybrid linear solvers for large 3D problems |
| title_full_unstemmed |
On the parallel scalability of hybrid linear solvers for large 3D problems |
| title_sort |
on the parallel scalability of hybrid linear solvers for large 3d problems |
| publishDate |
2008 |
| url |
http://oatao.univ-toulouse.fr/7711/1/haidar.pdf |
| work_keys_str_mv |
AT haidarazzam ontheparallelscalabilityofhybridlinearsolversforlarge3dproblems |
| _version_ |
1718553412246700032 |