Research Software
Parallel SpMV and SpMVT using CSB: A small Cilk++ library that performs sparse matrix times dense vector and sparse matrix transpose times dense vector using compressed sparse blocks.Source code
Associated paper
Combinatorial BLAS: Distributed memory reference implementation of the sparse linear algebra routines to support large scale graph analysis. Check back at the end of this summer (2009) for a release that uses MPI. In the meantime, you can check the design of PSpGEMM , a C++ library that performs sequential and parallel sparse matrix multiplication, primarily written to experimentally evaluate the performance of our algorithms.
Doxygen of PSpGEMM
PSpGEMM Wiki
GPU-APSP: A CUDA program that computes the distances for all pairs shortest paths in a dense directed graph (represented in adjacency matrix form).
Source code of GPU-APSP
Associated paper
© 2008 Aydın Buluç | Design by Andreas Viklund