Dear users,
I am trying to run Nektar on Windows using MPI and I am following the instructions in the blog post of May 18 2020.
I find that the code builds successfully and it runs the example (cyl.xml) using multiple threads but it looks like multiple copies of the same work are being run, rather than a single (faster) multi-threaded simulation.
I wonder if anyone has experienced the same issue, or has any advice on how to fix this? Could it be some kind of MPI version conflict (I imagine the `multiple copies' is the default behaviour if MPI is not working)?
Thanks,
Ed Threlfall.