Efficient Sparse Matrix-Vector Multiplication on CUDA Nathan Bell∗ and Michael Garland† December 11, 2008 Abstract The massive parallelism of graphics processing units (GPUs) offers tremendous performance in many - Band matrix - Document - PDFSEARCH.IO

First Page	Meta Content
	Efficient Sparse Matrix-Vector Multiplication on CUDA Nathan Bell∗ and Michael Garland† December 11, 2008 Abstract The massive parallelism of graphics processing units (GPUs) offers tremendous performance in many Add to Reading List Document Date: 2010-01-04 23:03:34 Open Document File Size: 3,44 MB Share Result on Facebook Company GPU / NVIDIA Corporation / / / IndustryTerm parallel device / sparse matrix-vector product / conventional multicore processors / computing / manycore processors / sparse linear systems / large-scale linear systems / / Person Michael Garland / / Position scheduler / COO / a hybrid ELL/COO / COO / remaining entries COO / programmer / / ProgrammingLanguage C / / Technology shared memory / manycore processors / conventional multicore processors / throughput-oriented processors / / SocialTag Sparse matrices Numerical linear algebra Matrix theory Matrices Arrays Sparse matrix Matrix representation Matrix Band matrix Algebra

Efficient Sparse Matrix-Vector Multiplication on CUDA Nathan Bell∗ and Michael Garland† December 11, 2008 Abstract The massive parallelism of graphics processing units (GPUs) offers tremendous performance in manyAdd to Reading List