Using sampled information: is it enough for the sparse matrix–vector product locality optimization? | Publicación