Compression and load balancing for efficient sparse matrix‐vector product on multicore processors and graphics processing units | Publicación