8 Hits in 4.9 sec

Improving the performance of parallel triangularization of a sparse matrix using a reconfigurable multicomputer [chapter]

José L. Sánchez, José M. García, Joaquin Fernández
1996 Lecture Notes in Computer Science  
In this paper we improve t he performance of the fast Givens rotations algorithm for sparse matrices by m eans of using a recon gurable multicomputer.  ...  In this paper we s h ow t he improvement for applying t his novel technique to improve t he performance of the parallel fast Givens rotations, and w e present g e n eral concepts related with it.  ...  In this paper, we present a n o vel technique to improve t he p e r f o r m ance of parallel triangularization of a sparse matrix using fast Givens rotations.  ... 
doi:10.1007/3-540-60902-4_52 fatcat:lwzvqt22qrbpdm6qs3k7ebxmp4

Dynamic reconfiguration of node location in wormhole networks

José L. Sánchez, José M. Garcı́a
2000 Journal of systems architecture  
Several techniques have been developed to increase the performance of parallel computers. Recon®gurable networks can be used as an alternative to increase the performance.  ...  We have used both a crossbar and a multistage interconnection network to implement a recon®gurable logical two-dimensional (2-D) torus topology.  ...  Both of them are described below. Triangularization of a sparse matrix We have used a parallel application such as the triangularization of a sparse matrix using Fast Givens [3] rotations.  ... 
doi:10.1016/s1383-7621(99)00044-2 fatcat:7teujzb3k5dfjjlqb6wrkfdtiu

Page 5104 of Mathematical Reviews Vol. , Issue 911 [page]

1991 Mathematical Reviews  
Plotkin and Eva Tardos, Improved dual network simplex (pp. 367-376); Carolyn Haibt Norton, Serge A. Plotkin and Eva Tardos, Using separation algorithms in fixed dimension (pp. 377-387). Michel X.  ...  In addition, when a given network contains one or more of the aforementioned iterative contact networks as subnetworks, the reliability analysis for the overall network can be simplified by using the formulas  ... 

Parallel LU factorization of sparse matrices on FPGA-based configurable computing engines

Xiaofang Wang, Sotirios G. Ziavras
2004 Concurrency and Computation  
In this paper, we describe our design and implementation of a parallel machine on an SOPC development board, using multiple instances of a soft IP configurable processor; we use this machine for LU factorization  ...  The intricacies of our FPGA-based design are presented along with tradeoff choices made for the purpose of illustration. Performance results prove the viability of our approach.  ...  Also, the paper has been improved substantially because of the suggestions made by the four referees.  ... 
doi:10.1002/cpe.748 fatcat:3n3dt37nabei7jdrlcfvoz72dq

3. A Bibliography on Parallel and Vector Numerical Algorithms [chapter]

James M. Ortega, Robert G. Voigt, Charles H. Romine
1990 Parallel Algorithms for Matrix Computations  
Since parallel and vector computation is expanding rapidly, we hope that the references we have collected over the years will be of some value to researchers entering the field.  ...  It will be used for research that has reached a certain level of maturity but needs additional refinement, for technical reviews or position statements, for bibliographies, and for computer software.  ...  Preliminary timing study f o r the CRA YPACK library, Internal Memorandum G4550-CM-39, Boeing Computer Services. [1982] . Improving the performance of a sparse matrix solver on the [1982] .  ... 
doi:10.1137/1.9781611971705.ch3 fatcat:pkjwqy26bfcjneyf76jqebybei

Parallel Circuit Simulation: A Historical Perspective and Recent Developments

Peng Li
2011 Foundations and Trends® in Electronic Design Automation  
author gain a deep understanding of circuit simulation.  ...  The author thanks Sani Nassif for collaborating on multialgorithm parallel circuit simulation.  ...  As a result, the parallelization of the triangular solves must use the same data distribution employed in the parallel factorization step.  ... 
doi:10.1561/1000000020 fatcat:knc64muljfhrrl5ynfsjauobpy

An algorithm for dynamic reconfiguration of a multicomputer network

J.M. Garcia, J. Duato
Proceedings of the Third IEEE Symposium on Parallel and Distributed Processing  
The dynamic reconfiguration of the interconnection network is an advanced feature of some multicomputers to reduce the communication overhead.  ...  That algorithm is transparent to the application programmer and is not restricted to a particular class of applications, being very well suited for parallel applications whose communication pattern varies  ...  A complex case: the sparse matrix triangularization.  ... 
doi:10.1109/spdp.1991.218232 dblp:conf/spdp/GarciaD91 fatcat:3n4ubp36h5fnjlmawuyib7gxoq


Ayham Zaza
2015 unpublished
After all, the inverse of a sparse matrix is a full matrix.  ...  Skyline Storage (SS) assumes a triangular matric, and traverses matrix elements column wise in upper triangular or row wise in lower triangular until it hits the diagonal.  ...