Parallel sparse matrix vector multiplication скачать в хорошем качестве

Parallel sparse matrix vector multiplication

This video shows how to turn a sequential algorithm for sparse matrix-vector multiplication (SpMV) into a parallel algorithm for a given data distribution. The resulting parallel SpMV has four supersteps: communication, computation, communication, computation. The distribution of the sparse matrix is shown in both a global view and a local view. I analyse the cost of the parallel algorithm under the assumption of equal spreading of matrix nonzeros and vector components. This video corresponds to Section 4.3 of the book Parallel Scientific Computation: A Structured Approach Using BSP, Second Edition, by Rob H. Bisseling, Oxford University Press, 2020. An expanded set of slides, solutions to the homework questions, and software accompanying the book can all be found on my personal book page: https://webspace.science.uu.nl/~bisse...