Filters








108 Hits in 1.4 sec

Electromechanical transients simulation on a multicomputer via the VDHN-Maclaurin method

F. Morales, H. Rudnick, A. Cipriano
2001 IEEE Transactions on Power Systems  
The parallel simulation was written in C language and implemented on a Parsytec PowerXplorer multicomputer.  ...  Such condition keeps a close relation with a physical coupling property exhibited by power systems, and the characteristics of the simulation method.  ...  Implementation on a Multicomputer The Parsytec PowerXplorer multicomputer consists of two PowerXplorer systems, each one with four PowerPC 601 RISC processors that work with 64 bits and 80 Mflops.  ... 
doi:10.1109/59.932277 fatcat:g3xgdpbzwzfw7katonq3gwy34m

Electromechanical Transients Simulation on a Multicomputer via the VDHN-Maclaurin Method

F. Morales, H. Rudnick, A. Cipriano
2001 IEEE Power Engineering Review  
The parallel simulation was written in C language and implemented on a Parsytec PowerXplorer multicomputer.  ...  Such condition keeps a close relation with a physical coupling property exhibited by power systems, and the characteristics of the simulation method.  ...  Implementation on a Multicomputer The Parsytec PowerXplorer multicomputer consists of two PowerXplorer systems, each one with four PowerPC 601 RISC processors that work with 64 bits and 80 Mflops.  ... 
doi:10.1109/mper.2001.4311439 fatcat:zn7inhoh7bgtrkgatupn2rvhxm

A programming methodology for dual-tier multicomputers

S.B. Baden, S.J. Fink
2000 IEEE Transactions on Software Engineering  
We support our arguments with empirical data from running the applications on various single-and dual-tier multicomputers.  ...  Hierarchically-organized ensembles of shared memory multiprocessors possess a richer and more complex model of locality than previous generation multicomputers with single processor nodes.  ...  LU Decomposition Finally, we consider the blocked right-looking distributed LU factorization algorithm of SCaLAPACK 30] .  ... 
doi:10.1109/32.842948 fatcat:mbs2jty2effw5dcvckrmulqlmq

LAPACK++

R. Pozo, J. J. Dongarra, D. W. Walker
1993 Proceedings of the 1993 ACM/IEEE conference on Supercomputing - Supercomputing '93  
Preliminary experiments of an object-based LU factorization algorithm using an SBS decomposition [2] suggest these algorithms scale well on multicomputers.  ...  Figure 2 2 illustrates performance characteristics of t,he LU factorizst,ion of various matrices on the same architecture using the LAPACE;++ LaLUFactorIP(A,F) routine, which overwrites A with it,s LU  ... 
doi:10.1145/169627.169680 dblp:conf/sc/DongarraPW93 fatcat:2xdi45tdtjcqti7mwx32h7wjv4

Page 1001 of Mathematical Reviews Vol. , Issue 93b [page]

1993 Mathematical Reviews  
Summary: “An LU decomposition of a centrosymmetric matrix, the Cholesky decomposition of a centrosymmetric, symmetric, and positive definite matrix, and a method of finding the inverse using Cholesky decomposition  ...  V. (6-NANA-S); Venkata Ramana, K. (6-NANA) On LU decomposition of a centrosymmetric matrix. Inform. Sci. 63 (1992), no. 1-2, 3-10.  ... 

PERFORMANCE STUDY OF LU FACTORIZATION WITH LOW COMMUNICATION OVERHEAD ON MULTIPROCESSORS

F. DESPREZ, J. J. DONGARRA, B. TOURANCHEAU
1995 Parallel Processing Letters  
In this paper, we make e cient use of asynchronous communications on the LU decomposition algorithm with pivoting and a column-scattered data decomposition to derive precise computational complexities.  ...  We then compare these results with experiments on the Intel iPSC/860 and Paragon machines and show that very good performances can be obtained on a ring with asynchronous communications.  ...  Experiments The experiments were performed on the Intel machines. On the iPSC/860, we used 2 to 64 processors with a ring embedded in the hypercube.  ... 
doi:10.1142/s012962649500014x fatcat:patgufmkkjhuth73moe45fikeq

High performance computing in power system applications [chapter]

Djalma M. Falcão
1997 Lecture Notes in Computer Science  
The paper starts with a brief introduction to the di erent types of HPC platforms adequate to power system applications.  ...  Finally, a critical review of the recent developed research work in the eld, along with prospective developments, is presented.  ...  , 16] is based on the AIS, the decomposition of the network equations in a Block Bordered Diagonal Form (BBDF), and a hybrid solution scheme using LU decomposition and the CG method.  ... 
doi:10.1007/3-540-62828-2_109 fatcat:2zmyyypiebed3a2rfogjbkrvde

Parallel Algorithm with Parameters Based on Alternating Direction for Solving Banded Linear Systems

Xinrong Ma, Sanyang Liu, Manyu Xiao, Gongnan Xie
2014 Mathematical Problems in Engineering  
An efficient parallel iterative method with parameters on distributed-memory multicomputer is investigated for solving the banded linear equations in this work.  ...  Numerical experiments implemented on HP rx2600 cluster verify that our algorithm has the advantages over the multisplitting one of high efficiency and low memory space, which has a considerable advantage  ...  multicomputers.  ... 
doi:10.1155/2014/752651 fatcat:dvieldwtdjamfaop2kwsm3mxpu

Alternating-Direction Line-Relaxation Methods on Multicomputers

Jörn Hofhaus, Eric F. Van de Velde
1996 SIAM Journal on Scientific Computing  
In our experiments, the problem size was determined by resolution requirements of the application.  ...  Our best results were obtained with a modified half-Gauss-Seidel line-relaxation method implemented by means of a new iterative block-tridiagonal solver that is developed here.  ...  With process boundaries that are parallel to the relaxation lines, the block-tridiagonal systems can be solved by sequential LU-decomposition, which is the optimal procedure for sequential computations  ... 
doi:10.1137/s1064827593253872 fatcat:mif5344sjvgavfmpxq36tbjvpy

Performance implications of communication mechanisms in all-software global address space systems

Beng-Hong Lim, Chi-Chao Chang, Grzegorz Czajkowski, Thorsten von Eicken
1997 SIGPLAN notices  
Regionbased caching benefits applications with irregular structure and with sufficient temporal locality, especially under high communication latencies.  ...  However, caching also hurts performance when there is insufficient data reuse or when the size of caching granularity is mismatched with the communication granularity.  ...  Acknowledgments We wish to thank Kirk Johnson for help with CRL, Eric Anderson for help with the Split-C version of Barnes-Hut, and Klaus Schauser for providing us access to the Meiko CS-2 at UCSB (acquired  ... 
doi:10.1145/263767.263794 fatcat:fvrtwyswrzhize766ey5z2fioa

Performance implications of communication mechanisms in all-software global address space systems

Beng-Hong Lim, Chi-Chao Chang, Grzegorz Czajkowski, Thorsten von Eicken
1997 Proceedings of the sixth ACM SIGPLAN symposium on Principles and practice of parallel programming - PPOPP '97  
Regionbased caching benefits applications with irregular structure and with sufficient temporal locality, especially under high communication latencies.  ...  However, caching also hurts performance when there is insufficient data reuse or when the size of caching granularity is mismatched with the communication granularity.  ...  Acknowledgments We wish to thank Kirk Johnson for help with CRL, Eric Anderson for help with the Split-C version of Barnes-Hut, and Klaus Schauser for providing us access to the Meiko CS-2 at UCSB (acquired  ... 
doi:10.1145/263764.263794 dblp:conf/ppopp/LimCCE97 fatcat:praoazuybfga7awohqjwxb4qxq

An Efficient Parallel Algorithm to Solve Block?Toeplitz Systems

P. Alonso, J. M. Bad�a, A. M. Vidal
2005 Journal of Supercomputing  
In this paper, we present an efficient parallel algorithm to solve Toeplitz-block and block-Toeplitz systems in distributed memory multicomputers.  ...  We obtain the LU decomposition of M by repeating this process as shown in Algorithm 1.  ...  If the generators are in the proper form, then S − F S A T = [G −u Fu] [B v Av] T = [Fu G 1:n,2:r ] [Av B 1:n,2:r ] T = G B T . (10) The Schur algorithm computes the LU decomposition of matrix M, with  ... 
doi:10.1007/s11227-005-0182-6 fatcat:wz5ur4ei5vb23azoinladnmxza

Automatic selection of dynamic data partitioning schemes for distributed-memory multicomputers [chapter]

Daniel J. Palermo, Prithviraj Banerjee
1996 Lecture Notes in Computer Science  
of dynamic decomposition.  ...  E v a lu a tio n In order to evaluate the effectiveness of dynamic distribution, the ADI program with a problem size of 512 x 5126 is compiled with both a fully static distribution (one iteration shown  ... 
doi:10.1007/bfb0014213 fatcat:kuboyq3lqvesrfgfbgrjsvpdsm

Porting industrial codes and developing sparse linear solvers on parallel computers

Michel J. Dayde, Iain S. Duff
1995 Computing Systems in Engineering  
We report in some detail on the parallelization We illustrate this by using examples from our experience in porting industrial codes and and efficient parallel code. of computers).  ...  may one day be capable of generating efficient parallel code for is small compared with the computation. from our experience, it is restricted to applications where the amount of communication environments  ...  Table 10 : 10 Performance in MFlops of the LU factorization on shared memory Computer Procs I Serial I Parallel BLAS I Parallel LU I Routine I Performance Block LU factorization Manufacturer's Library  ... 
doi:10.1016/0956-0521(95)00033-x fatcat:fwvkd3wkdvagrjdhra7latjcme

AN ASSESSMENT OF THE CONNECTION MACHINE

ROBERT SCHREIBER
1993 International journal of high speed computing  
These are contrasted with the same issues in MD multiprocessors and multicomputers.  ...  There are several well known examples: the 5Gflop matrix multiplier, the TMC implementation of FFT, Vavasis' fast spreads for the QR factorization, the LU decomposition codes in the new linear algebra  ...  In both classes of machine, hardware costs are roughly the same, with slightly less hardware devoted to interconnect in the multicomputers. In early multicomputers, memory per node was inadequate.  ... 
doi:10.1142/s0129053393000220 fatcat:simo6to2hnamhec4itephqirvm
« Previous Showing results 1 — 15 out of 108 results