I was not aware that SolverCG is using multiple threads. I even briefly scanned the source code, but could not spot anything that looked like it.
It's not SolverCG itself but things like SparseMatrix::vmult that are parallelized. If you have a simple preconditioner you spend much time in vmult, so you get decent speedup.
Best W. ------------------------------------------------------------------------ Wolfgang Bangerth email: [email protected] www: http://www.math.tamu.edu/~bangerth/ _______________________________________________ dealii mailing list http://poisson.dealii.org/mailman/listinfo/dealii
