A NUMA Aware Scheduler for a Parallel Sparse Direct Solver
FAVERGE, Mathieu
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Efficient runtime systems for parallel architectures [RUNTIME]
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Efficient runtime systems for parallel architectures [RUNTIME]
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
LACOSTE, Xavier
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
RAMET, Pierre
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
FAVERGE, Mathieu
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Efficient runtime systems for parallel architectures [RUNTIME]
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Efficient runtime systems for parallel architectures [RUNTIME]
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
LACOSTE, Xavier
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
RAMET, Pierre
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
< Reduce
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
Language
en
Communication dans un congrès
This item was published in
PMAA'08, 2008, Neuchâtel. 2008
English Abstract
Over the past few years, parallel sparse direct solvers made significant progress and are now able to solve efficiently industrial three-dimensional problems with several millions of unknowns. An hybrid MPI-thread ...Read more >
Over the past few years, parallel sparse direct solvers made significant progress and are now able to solve efficiently industrial three-dimensional problems with several millions of unknowns. An hybrid MPI-thread implementation of our direct solver PaStiX is already well suited for SMP nodes or new multi-core architectures and drastically reduced the memory overhead and improved scalability. In the context of distributed NUMA architectures, a dynamic scheduler based on a work-stealing algorithm has been developed to fill in communication idle times. On these architectures, it is important to take care of NUMA effects and to preserve memory affinity during the work-stealing. The scheduling of communications also needs to be adapted, especially to ensure the overlap by computations. Experiments on numerical test cases will be presented to prove the efficiency of the approach on NUMA architectures. If memory is not large enough to treat a given problem, disks must be used to store data that cannot fit in memory (out-of-core storage). The idle-times due to disk access have to be managed by our dynamic scheduler to prefetch and save datasets. Thus, we design and study specific scheduling algorithms in this particular context.Read less <
English Keywords
sparse direct solver
NUMA architecture
multi-cores
dynamic scheduling
ANR Project
Adaptation et Optimisation des Performances Applicatives sur architectures NUMA. Etude et Mise en Œuvre sur des Applications en SISmologie. - ANR-05-CIGC-0002
Origin
Hal imported