Filters








5 Hits in 6.4 sec

Rectangular Full Packed Format for Cholesky's Algorithm: Factorization, Solution and Inversion [article]

Fred G. Gustavson, Jerzy Wasniewski, Jack J. Dongarra, Julien Langou
2009 arXiv   pre-print
via the use of Level 3 BLAS.  ...  The standard two dimensional arrays of Fortran and C (also known as full format) that are used to represent triangular and symmetric matrices waste nearly half of the storage space but provide high performance  ...  Hansen for consulting on the IBM and SGI systems; and Bjarne Stig Andersen for obtaining the results on the Itanium and NEC computers.  ... 
arXiv:0901.1696v1 fatcat:adjikfgaqfdotmzerrvzalywfi

Recursive Blocked Algorithms and Hybrid Data Structures for Dense Matrix Library Software

Erik Elmroth, Fred Gustavson, Isak Jonsson, Bo Kågström
2004 SIAM Review  
This article reviews and details some of the recent advances made by applying the paradigm of recursion to dense matrix computations on today's memory-tiered computer systems.  ...  The results we survey include new algorithms and library software implementations for level 3 kernels, matrix factorizations, and the solution of general systems of linear equations and several common  ...  Especially, we thank Per Ling, former member of the Umeå team; the former Master students André Henriksson, Olov Gustavsson, and Andreas Lindkvist; and Bjarne Andersen and Jerzy Wasniewski of UNI-C, Lyngby  ... 
doi:10.1137/s0036144503428693 fatcat:7zmqj5eee5adxk56lbccrlyq3m

Efficient solution of sparse linear systems arising in engineering applications on vector hardware [article]

Sunil Reddy Tiyyagura, Universität Stuttgart, Universität Stuttgart
2010
These studies show that the vector systems are well balanced than most scalar systems with respect to many aspects that determine the sustained performance of many real world applications.  ...  The approach adopted in this work reduces the load on the memory subsystem by blocking all the unknowns at each mesh point and then solving the resulting blocked global system of equations.  ...  The number of iterations needed for convergence in BLIS for each newton step varies largely between 200-2000 depending on the problem size (number of equations).  ... 
doi:10.18419/opus-1878 fatcat:72ce3l7dojaphdzyecqsmdcopy

The Distribution of the MusselMytilusSpecies Along the Norwegian Coast

Steven J. Brooks, Eivind Farmen
2013 Journal of Shellfish Research  
Inject crude extract in 200-2000 µL batches, depending on the capacity of the GPC column (semi prep 25 mm column may be loaded with 2000 µL).  ...  Test Name of the test reference Reference to the origin of the data year Year of production Country lab Laboratory that performed the analyses type Is it a control or other type of sample  ...  Secretariat facilities: About [one month] of the services of Secretariat Professional and General Staff will be required. Financial: Cost of production and publication of a 170-page CRR.  ... 
doi:10.2983/035.032.0203 fatcat:ozwkkrxbibbk3pwh73es7yaqm4

The evolution of the PVM concurrent computing system

G.A. Geist, V.S. Sunderam
Digest of Papers. Compcon Spring  
The results re-l_orted indicate that MD simulations for sinall systems As mentiolmd earlier, several hundred sites are ac-('200-2000 atoms) require approximately equal times tively using PVM at the tiine  ...  Fig-use of PVM constructs and host language col_trol flow tire 1 depicts a simplified architectural overview of the statements. PVM system.  ... 
doi:10.1109/cmpcon.1993.289733 fatcat:ptkx3gae4fb4tlvsdiziiskn3e