| Document Date: 2012-05-29 06:13:35
Open Document File Size: 350,68 KBShare Result on Facebook
City Vienna / / Company GPU / Hybrid GPU Accelerated Manycore Systems / AMD / NVIDIA / INTEL / / Country Austria / / Currency AMD / / Facility ViennaCL The Vienna Computing Library / GotoBLAS2 library / C++ Standard Template Library / libflame Library / ViennaCL library / SimuNova Matrix Template Library / Austria Institute / CUDA library / NVIDIAs CUDA library / Matrix Template Library / / IndustryTerm specific technology / alone computing clusters / memory algorithm / distributed memory systems / matrix-matrix product / specific computing target / matrix product algorithms / matrix product algorithm / cluster systems / matrix products / consumer level computing / dense matrix-matrix product / large dense matrix-matrix products / respective computing nodes / multi-processor / heterogenous computing nodes / computing / scientific computing / workstation solutions / / OperatingSystem Linux / / Organization Institute for Microelectronics / European Research Council / Austria Institute for Analysis / Graduate School / US Federal Reserve / / Person Ai / Karl Rupp / / Product i7 960 4C/8T 12 / Phenom II X4 965 4C/4T 8 / Opteron / / ProgrammingLanguage C++ / / PublishedMedium Journal of Physics / Technology Linux / API / Hyperthreading technology / memory algorithm / partitioning algorithm / ViennaCL matrix product algorithm / Parallel Processing / specific technology / ViennaCL matrix product algorithms / / URL http /
SocialTag |
Towards Distributed Heterogenous High-Performance Computing with ViennaCL Josef Weinbub1 , Karl Rupp1,2 , and Siegfried Selberherr1 1 2