| http://www.w3.org/ns/prov#value | - Thus, such parallel algorithm is more suitable for computers of shared memory systems, and an OpenMP-based platform is employed.The parallelization procedures for the PCG solver can be summarized as follows:(i)first, the matrix and vector operations are parallelized, including and , where the vector is generated by specific algorithm;(ii)then, the inner product operation is parallelized;(iii)furth
|
