×

OSKI

swMATH ID: 7006
Software Authors: Berkeley Benchmarking and OPtimization (BeBOP) Project
Description: OSKI: Optimized Sparse Kernel Interface. What is OSKI? The Optimized Sparse Kernel Interface (OSKI) Library is a collection of low-level C primitives that provide automatically tuned computational kernels on sparse matrices, for use in solver libraries and applications. OSKI has a BLAS-style interface, providing basic kernels like sparse matrix-vector multiply and sparse triangular solve, among others. The current implementation targets cache-based superscalar uniprocessor machines, though we are developing extensions for vector architectures, SMPs, and large-scale distributed memory machines.
Homepage: http://bebop.cs.berkeley.edu/oski/
Programming Languages: C
Keywords: sparse matrices
Related Software: ATLAS; SparseMatrix; FFTW; Trilinos; LAPACK; Petabricks; SPARSKIT; PETSc; SPIRAL; CUDA; JDQZ; JDQR; BLAS; PHiPAC; PAPI; ADAPT; OpenCL; Sparsity; MKL; SBR Toolbox
Cited in: 21 Publications

Citations by Year