Filters








951 Hits in 5.6 sec

Dynamic Load Balancing Strategies for Hierarchical p-FEM Solvers [chapter]

Ralf-Peter Mundani, Alexander Düster, Jovana Knežević, Andreas Niggl, Ernst Rank
2009 Lecture Notes in Computer Science  
To fully exploit the underlying computing power of parallel systems, dynamic load balancing strategies become an essential component.  ...  Applying hierarchical concepts based on a nested dissection approach allow for both the design of sophisticated solvers as well as for advanced parallelisation strategies.  ...  Conclusion In this paper, we have proposed a dynamic load balancing strategy for linear equation solvers based on the nested dissection approach in order to tackle known problems related to tree parallelisations  ... 
doi:10.1007/978-3-642-03770-2_37 fatcat:3jhsmp57s5hqljieytnqldlk6a

A Static Parallel Multifrontal Solver for Finite Element Meshes [chapter]

Alberto Bertoldo, Mauro Bianco, Geppino Pucci
2006 Lecture Notes in Computer Science  
To balance the load among the processors, we devise a simple model-driven partitioning strategy to precompute a high-quality balancing for a large family of structured meshes.  ...  The resulting approach is proved to be considerably more efficient than the strategies implemented by MUMPS and SuperLU DIST, two state-of-the-art parallel multifrontal solvers.  ...  MUMPS implements a multifrontal algorithm based on a parallel hierarchical LU decomposition approach similar to the one used in our solver, but it follows a dynamic approach to distribute the work between  ... 
doi:10.1007/11946441_67 fatcat:w2hbrt2blfhhzasj3pimh26un4

Data structures and load balancing for parallel adaptive hp finite-element methods

A.K. Patra, A. Laszloffy, J. Long
2003 Computers and Mathematics with Applications  
Adaptive hp finite-element methods (FEM)  ...  for parallel adaptive meshing, l the use of predictive strategies for dynamic load balancing that greatly reduce excessive data migration during rebalancing of load among processors, and finally, l  ...  Several dynamic load-balancing schemes designed to obtain good load balance while minimizing data migration have been presented.  ... 
doi:10.1016/s0898-1221(03)90084-8 fatcat:kfbmcn4sk5eptguihyc5jhmqbm

FEMS: An Adaptive Finite Element Solver

Alberto Bertoldo
2007 2007 IEEE International Parallel and Distributed Processing Symposium  
FEMS relies on a simple model-driven mesh partitioning strategy, which makes it possible to perform efficient static load balancing on both homogeneous and heterogeneous machines.  ...  We present the software architecture of FEMS, a parallel multifrontal solver for FE applications whose main feature is an install-time training phase where adaptation to the computing platform takes place  ...  perform static load balancing.  ... 
doi:10.1109/ipdps.2007.370569 dblp:conf/ipps/Bertoldo07 fatcat:gn3biypjwndg3hb7elkecelwwm

Computational Steering of Complex Flow Simulations [chapter]

Atanas Atanasov, Hans-Joachim Bungartz, Jérôme Frisch, Miriam Mehl, Ralf-Peter Mundani, Ernst Rank, Christoph van Treeck
2010 High Performance Computing in Science and Engineering, Garching/Munich 2009  
The gap between steering and HPC is bridged by a hierarchical approach that performs fast interactive simulations for many scenario variants increasing the accuracy via hierarchical refinements in dependence  ...  The environment consists of efficient fluid dynamics simulation codes and a steering and visualisation framework providing a user interface, communication methods for distributed steering, and parallel  ...  Adaptive Load Balancing Within a related project for structural analysis using the p-version finite element method (p-FEM, [12] ) -i. e. increasing the polynomial degree p of the shape functions for better  ... 
doi:10.1007/978-3-642-13872-0_6 fatcat:xaou7c27urbhbasxwtfuatrkny

Hierarchical Load Balancing for Charm++ Applications on Large Supercomputers

Gengbin Zheng, Esteban Meneses, Abhinav Bhatele, Laxmikant V. Kale
2010 2010 39th International Conference on Parallel Processing Workshops  
We show performance data of the hierarchical load balancing method on up to 16,384 cores of Ranger (at TACC) for a synthetic benchmark.  ...  In this paper, we present an automatic dynamic hierarchical load balancing method that overcomes the scalability challenges of centralized schemes and poor solutions of traditional distributed schemes.  ...  Comparison of time to solution for centralized and hierarchical load balancers for NAMD on Blue Gene/P load balancing is done according to the semi-centralized load balancing scheme described in Section  ... 
doi:10.1109/icppw.2010.65 dblp:conf/icppw/ZhengMBK10 fatcat:ikr23gspcrgnlf4jkmsltbpgwe

A Parallel Sparse Direct Solver via Hierarchical DAG Scheduling

Kyungjoo Kim, Victor Eijkhout
2014 ACM Transactions on Mathematical Software  
Kim and Eijkhout Parallel Sparse Direct Solver via Hierarchical DAG Scheduling 1.  ...  The advanced hp-FEM typically provides an order of magnitude higher solution resolution than conventional linear FEM by using both variable element size (h) and higher order of approximation (p).  ...  Codes are available under the GNU Lesser General Public License (LGPL) for the non-commercial use at http://code.google.com/p/uhm.  ... 
doi:10.1145/2629641 fatcat:qbiygwm4b5bi7b2mtp6mm323qe

Periodic hierarchical load balancing for large supercomputers

Gengbin Zheng, Abhinav Bhatelé, Esteban Meneses, Laxmikant V. Kalé
2011 The international journal of high performance computing applications  
We show performance data of the hierarchical load balancing method on up to 16, 384 cores of Ranger cluster (at TACC) and 65, 536 cores of a Blue Gene/P at Argonne National Laboratory for a synthetic benchmark  ...  In this paper, we present an automatic dynamic hierarchical load balancing method that overcomes the scalability challenges of centralized schemes and poor solutions of traditional distributed schemes.  ...  We used running time on the Blue Gene/P at Argonne National Laboratory, which is supported by DOE under contract  ... 
doi:10.1177/1094342010394383 fatcat:n6aqh2atfzd3zd3aegkjdg2jxy

Modular FEM framework "ModFem" for generic scientific parallel simulations

Michalik Kazimierz, Banas Krzysztof, Plaszewski Przemyslaw, Cybulka Pawel
2013 Computer Science  
We present the design for, and implementation of, a flexible and robust parallel modular finite element (FEM) framework called ModFEM.  ...  The structure allows for reusing the sequential code for parallel environments, and also supports solving multi-physics and multi-scale problems.  ...  Domain decomposition and further dynamic load balancing were also the subject of research and articles.  ... 
doi:10.7494/csci.2013.14.3.513 fatcat:632rfec2tjg7fgsnssk4jfeyli

An agent-oriented hierarchic strategy for solving inverse problems

Maciej Smołka, Robert Schaefer, Maciej Paszyński, David Pardo, Julen Álvarez-Aramberri
2015 International Journal of Applied Mathematics and Computer Science  
The strategy offers exceptionally low computational costs, mainly because the direct solver accuracy (performed by the hp-adaptive finite element method) is dynamically adjusted for each inverse search  ...  The paper discusses the complex, agent-oriented hierarchic memetic strategy (HMS) dedicated to solving inverse parametric problems.  ...  During the computations, An agent-oriented hierarchic strategy for solving inverse problemsFig. 4. Domain geometry.  ... 
doi:10.1515/amcs-2015-0036 fatcat:hsmgb6bqmnewnjpmttzrb6lqha

A High-Performance Interactive Computing Framework for Engineering Applications [chapter]

Jovana Knežević, Ralf-Peter Mundani, Ernst Rank
2013 Lecture Notes in Computational Science and Engineering  
To obtain rapid, albeit approximate feedback from the simulation in case of perpetual user interaction, a multi-hierarchical approach is advantageous.  ...  Namely, the regular course of the simulation coupled to our framework is interrupted in small, cyclic intervals followed by a check for updates.  ...  Chris Johnson for inviting her and all the researchers for fruitful discussions.  ... 
doi:10.1007/978-3-642-38762-3_9 fatcat:rzfsctjperdete4hwo4dgnlogy

A scalable H-matrix approach for the solution of boundary integral equations on multi-GPU clusters [article]

Helmut Harbrecht, Peter Zaspel
2018 arXiv   pre-print
On 1024 GPUs of the cluster Titan, it takes less than 6 minutes to solve the 1.5 million unknowns problem, with 5.7 minutes for the setup phase and 20 seconds for the iterative solver.  ...  Using a model GPU implementation of a boundary element method (BEM) solver, we are able to achieve more than 67 percent relative parallel speed-up going from 128 to 1024 GPUs for a model geometry test  ...  Similarly, we balance the storage size of the batched low-rank factors for the ACA task list, compare Figure 3 . Scalability and load balancing discussion.  ... 
arXiv:1806.11558v1 fatcat:dtn556nsrfe3hflgr7bygngfha

Implementation and Evaluation of an AMR Framework for FDM Applications

Masaharu Matsumoto, Futoshi Mori, Satoshi Ohshima, Hideyuki Jitsumoto, Takahiro Katagiri, Kengo Nakajima
2014 Procedia Computer Science  
In the AMR framework, a dynamic domain deco mposition (DDD) technique, as a dynamic load balancing method, is also implemented to correct the computational load imbalance between each process associated  ...  By performing a 3D AM R test simulat ion, it is confirmed that dynamic load balancing can be achieved and execution time can be reduced by introducing the DDD technique.  ...  Acknowledgement This work is funded by JST (Japan Science and Technology Agency)/CREST (Core Research for Evolutional Science and Technology).  ... 
doi:10.1016/j.procs.2014.05.084 fatcat:xa7dz3pcj5badnsboqdbdjs66m

A hierarchical sequential ALE poromechanics model for tire-soil-water interaction on fluid-infiltrated roads

Ines Wollny, WaiChing Sun, Michael Kaliske
2017 International Journal for Numerical Methods in Engineering  
HIERARCHICAL SEQUENTIAL ALE MODEL FOR TIRE-SOIL-WATER INTERACTION 913 With x = ALE + and ∇ ALE = 0, Equation (9) reads v (x, t) = ALE HIERARCHICAL SEQUENTIAL ALE MODEL FOR TIRE-SOIL-WATER INTERACTION 915  ...  This paper introduces a hierarchical sequential arbitrary Lagrangian-Eulerian (ALE) model for predicting the tire-soil-water interaction at finite deformations.  ...  Including the staggered solid-fluid iteration scheme for the poromechanical ALE formulation into the staggered tire-pavement computation scheme demands for a hierarchical solution strategy, where the tire-pavement  ... 
doi:10.1002/nme.5537 fatcat:6jybqnzlyfgxdiopmjc3ztgyh4

Fast Multipole Preconditioners for Sparse Matrices Arising from Elliptic Equations [article]

Huda Ibeid, Rio Yokota, Jennifer Pestana, David Keyes
2016 arXiv   pre-print
Among optimal hierarchical algorithms for the computational solution of elliptic problems, the Fast Multipole Method (FMM) stands out for its adaptability to emerging architectures, having high arithmetic  ...  We demonstrate that, beyond its traditional use as a solver in problems for which explicit free-space kernel representations are available, the FMM has applicability as a preconditioner in finite domain  ...  We thank Lorena Barba for the support in developing Ex-aFMM.  ... 
arXiv:1308.3339v4 fatcat:skniridmdvfolbfpbwhga3wnl4
« Previous Showing results 1 — 15 out of 951 results