Re: [ESPResSo-devel] LB fluid with variable viscosity

[Michael Kuron]

Hi Ivan,

> Assume I can determin the intersection of blood cell boundaries and
> horizontal LB grid lines (by an algorithm based on a loop over the
> particles). I need to send this data from MPI slave to MPI master.
> But on each slave, I could have variable number of intersections. How
> can I implement this? Can I use the
> mpi_gather_stats/mpi_gather_stats_slave functions? How do I implement
> the variable size od data that need to be gathered?

As far as I understand, only neighboring processes need to exchange
data. So the communication shouldn't all go through MPI rank 0 (the
"master" rank). Besides, don't processes have ghost copies of blood
cells that intersect them but actually reside on neighboring blocks? If
that is the case, you don't need any communication; just calculate the
intersections on all ranks using whatever real and ghost particles are
there.

Michael