Filters








16,445 Hits in 5.6 sec

A parallel iterative linear system solver with dynamic load balancing

Peter Christen
2000 ANZIAM Journal  
This paper describes the design and implementation of a parallel iterative linear system solver for distributed memory multicomputers and workstation clusters.  ...  It is capable of applying heterogeneous data distribution and dynamic load balancing within an iterative solver routine at matrix level.  ...  Object oriented approaches try to hide the load balancing mechanism within their distributed objects. The aim behind our work is to provide a parallel iterative linear system solver with dlb.  ... 
doi:10.21914/anziamj.v42i0.605 fatcat:7zkn6q5u4ncifabqk374wbnep4

A parallel iterative linear system solver with dynamic load balancing

Peter Christen
1998 Proceedings of the 12th international conference on Supercomputing - ICS '98  
This paper describes the design and implementation of a parallel iterative linear system solver for distributed memory multicomputers and workstation clusters.  ...  It is capable of applying heterogeneous data distribution and dynamic load balancing within an iterative solver routine at matrix level.  ...  Object oriented approaches try to hide the load balancing mechanism within their distributed objects. The aim behind our work is to provide a parallel iterative linear system solver with dlb.  ... 
doi:10.1145/277830.277836 dblp:conf/ics/Christen98 fatcat:ex7kgfd2qbhetnhxskjmcdx23u

Development of a parallel three-phase transient stability simulator for power systems

Shrirang Abhyankar, Alexander Flueck, Xu Zhang, Hong Zhang
2011 Proceedings of the first international workshop on High performance computing, networking and analytics for the power grid - HiPCNA-PG '11  
The performance of the proposed parallel three-phase transient stability simulator using Krylov subspace based iterative solver GMRES with two variants of Block-Jacobi preconditioner and a very dishonest  ...  Unlike the existing transient stability simulators that use a balanced per phase transmission network, the authors use a three phase transmission network.  ...  Results obtained by using iterative solver GMRES, with two preconditioning schemes, are presented for three large-scale systems.  ... 
doi:10.1145/2096123.2096130 fatcat:lh6kwrmwknhjfjpx7at23drswa

Computational Challenges for Power System Operation

Yousu Chen, Zhenyu Huang, Yan Liu, Mark J. Rice, Shuangshuang Jin
2012 2012 45th Hawaii International Conference on System Sciences  
This paper presents some research outcomes in the areas of parallel state estimation using the preconditioned conjugated gradient method, parallel contingency analysis with a dynamic load balancing scheme  ...  and distributed system architecture.  ...  A parallel contingency analysis package with a counter-based dynamic load balancing scheme can obtain a nearly linear speed-up.  ... 
doi:10.1109/hicss.2012.171 dblp:conf/hicss/ChenHLRJ12 fatcat:q52cc6vgdvbhvfwjpr2xmpipbq

Dynamic Load Balancing of Parallel Computational Iterative Routines on Platforms with Memory Heterogeneity [chapter]

David Clarke, Alexey Lastovetsky, Vladimir Rychkov
2011 Lecture Notes in Computer Science  
This paper presents a new data partitioning algorithm that applies multidimensional solvers to numerical solution of the system of non-linear equations formalizing the problem of optimal data partitioning  ...  The better accuracy of the approximation in its turn results in a more optimal distribution of the computational load between the heterogeneous processors.  ...  Convergence and Complexity Dynamic Load Balancing of Parallel Computational Iterative Routines In this section, we demonstrate how the new data partitioning algorithm can be used for dynamic load balancing  ... 
doi:10.1007/978-3-642-21878-1_6 fatcat:h4hlt2ib7ngl3aorgcoozqlm5a

Using Multidimensional Solvers for Optimal Data Partitioning on Dedicated Heterogeneous HPC Platforms [chapter]

Vladimir Rychkov, David Clarke, Alexey Lastovetsky
2011 Lecture Notes in Computer Science  
This paper presents a new data partitioning algorithm that applies multidimensional solvers to numerical solution of the system of non-linear equations formalizing the problem of optimal data partitioning  ...  The better accuracy of the approximation in its turn results in a more optimal distribution of the computational load between the heterogeneous processors.  ...  Convergence and Complexity Dynamic Load Balancing of Parallel Computational Iterative Routines In this section, we demonstrate how the new data partitioning algorithm can be used for dynamic load balancing  ... 
doi:10.1007/978-3-642-23178-0_29 fatcat:mdamr2zfkbhodiwfcugtpcbmv4

Accelerating subsurface transport simulation on heterogeneous clusters

Oreste Villa, Nitin Gawande, Antonino Tumeo
2013 2013 IEEE International Conference on Cluster Computing (CLUSTER)  
Then, we combine the batched GPU solver with an OpenMP-based CPU solver, and present an adaptive load balancer that dynamically distributes the linear systems to solve between the two components inside  ...  This technique involves computing a Jacobian matrix and a residual vector for each set of equations, and then solving iteratively the linearized system by performing Gaussian Elimination and LU decomposition  ...  Then, we combine the batched GPU solver with an OpenMP-based CPU solver, and present an adaptive load balancer that dynamically distributes the linear systems to solve between the two components inside  ... 
doi:10.1109/cluster.2013.6702656 dblp:conf/cluster/VillaGT13 fatcat:jaculgjbdveclkj7abv7oyzkxy

Solvers for $\mathcal{O} (N)$ Electronic Structure in the Strong Scaling Limit

Nicolas Bock, Matt Challacombe, Laxmikant V. Kalé
2016 SIAM Journal on Scientific Computing  
Employing classic technologies associated with generalized N -Body solvers, including over-decomposition, recursive task parallelism, orderings that preserve locality, and persistence-based load balancing  ...  Comput. 35 C72, 2013], and involves a recursive, task-parallel algorithm, often employed by generalized N -Body solvers, to occlusion and culling of negligible products in the case of matrices with decay  ...  Dynamic persistence-based load balancing strategies migrate chares transparently during solver execution based on load and communication measurements from previous solver iterations and efficiently optimize  ... 
doi:10.1137/140974602 fatcat:vwyzeiwsazchtcfafbibq4sdt4

Real-Time Power System Dynamics Simulation Using a Parallel Block-Jacobi Preconditioned Newton-GMRES Scheme

Shrirang Abhyankar, Alexander J. Flueck
2012 2012 SC Companion: High Performance Computing, Networking Storage and Analysis  
We present a parallel linear solution scheme using Krylov subspace based iterative solver GMRES with a Block-Jacobi preconditioner that shows promising prospect of a realtime dynamics simulation.  ...  The main bottleneck in these simulations is the solution of the linear system during each nonlinear iteration of the Newton's method.  ...  A preconditioner is a matrix which transforms the linear system Ax = b into another system with a better spectral properties for the iterative solver.  ... 
doi:10.1109/sc.companion.2012.48 dblp:conf/sc/AbhyankarF12 fatcat:47xinaa77vc67ff5ill3sj4dlu

Page 5226 of Mathematical Reviews Vol. , Issue 98H [page]

1998 Mathematical Reviews  
(electronic)); Nasir Touheed and Peter Jimack, Dynamic load-balancing for adaptive PDE solvers with hierarchical refinement (8 pp. (elec- tronic)); A. A. Mirin, R. H. Cohen, W. P. Dannevik, A. M.  ...  Weiss [Riidiger Weiss}, LINSOL, a parallel iterative linear solver package of gen- eralized CG-type for sparse matrices (8 pp.  ... 

Enabling Next-Generation Parallel Circuit Simulation with Trilinos [chapter]

Chris Baker, Erik Boman, Mike Heroux, Eric Keiter, Siva Rajamanickam, Rich Schiek, Heidi Thornquist
2012 Lecture Notes in Computer Science  
These generally involve a higher-level partitioning of the devices [6] or lower-level partitioning of the linear system of equations [7] to facilitate the creation of a more efficient parallel matrix solver  ...  With the move towards multi-core CPUs and GPU technology, retaining this scalability on future parallel architectures will be a challenge.  ...  The approximationS is used as a preconditioner to iteratively solve the Schur complement system, with a relative residual tolerance of 10 −10 in at most 30 iterations.  ... 
doi:10.1007/978-3-642-29737-3_36 fatcat:hrkojy2vnrfndnywuumhh376xu

Solvers for O (N) Electronic Structure in the Strong Scaling Limit [article]

Nicolas Bock and Matt Challacombe and Laxmikant V. Kalé
2015 arXiv   pre-print
Employing classic technologies associated with generalized N-Body solvers, including over-decomposition, recursive task parallelism, orderings that preserve locality, and persistence-based load balancing  ...  Comput. 35 C72, 2013], and involves a recursive, task-parallel algorithm, often employed by generalized N-Body solvers, to occlusion and culling of negligible products in the case of matrices with decay  ...  Dynamic persistence-based load balancing strategies migrate chares trans- parently during solver execution based on load and communication measurements from previous solver iterations and efficiently optimize  ... 
arXiv:1403.7458v7 fatcat:75dqzyl2xbflhdvh5x6zsmfasq

Accelerating the computation of critical eigenvalues with parallel computing techniques

Petros Aristidou, Gabriela Hug
2016 2016 Power Systems Computation Conference (PSCC)  
In this paper, a procedure is proposed that uses domain decomposition and parallel computing methods, to accelerate the computation of eigenvalues in a selected region of the complex plane with iterative  ...  The proposed algorithm is validated on a small transmission system and its performance is assessed on a large-scale, combined transmission and distribution system.  ...  Thus, the dynamic strategy has been chosen for better load balancing.  ... 
doi:10.1109/pscc.2016.7540976 dblp:conf/pscc/AristidouH16 fatcat:lrpfnnjbxzbopahyazk4xangkm

Heuristic static load-balancing algorithm applied to the fragment molecular orbital method

Yuri Alexeev, Ashutosh Mahajan, Sven Leyffer, Graham Fletcher, Dmitri G. Fedorov
2012 2012 International Conference for High Performance Computing, Networking, Storage and Analysis  
Although dynamic load balancing is widespread, it is increasingly difficult to implement effectively with thousands of processors or more, prompting a second look at static load-balancing techniques even  ...  We propose a heuristic static load-balancing algorithm, employing fitted benchmarking data, as an alternative to dynamic load balancing.  ...  Mazanetz from Evotec for providing the PDB structure of Aurora-A kinase system used in our calculations.  ... 
doi:10.1109/sc.2012.62 dblp:conf/sc/AlexeevMLFF12 fatcat:ed4cztal7nhp7ddqnti4ypsvuq

Parallel programming with message passing and directives

S.W. Bova, C.P. Breshears, H. Gabb, B. Kuhn, B. Magro, R. Eigenmann, G. Gaertner, S. Salvini, H. Scott
2001 Computing in science & engineering (Print)  
These applications have three primary goals: • high speedup, scalable performance, and efficient system use; • similar behavior on a wide range of platforms and easy portability between platforms; and  ...  The COMPUTING IN SCIENCE & ENGINEERING PARALLEL PROGRAMMING WITH MESSAGE PASSING AND DIRECTIVES The authors discuss methods for expressing and tuning the performance of parallel programs, using two programming  ...  Iterative solvers, for example, consist of a parallel region with several loops. • Dynamic load balanced-In some applications, static scheduling results in gross inefficiency.  ... 
doi:10.1109/5992.947105 fatcat:of7laitsjnhz7exipyqhf7ehtq
« Previous Showing results 1 — 15 out of 16,445 results