<div dir="ltr">As Jed says hyperbolic dominated systems are hard. <div><br></div><div>For geometric papers you can look at Jameson papers and Adams, Samtaney & Brandt (MHD).<div><br></div><div>But be prepared to work. Good stuff but hard.</div></div><div><br></div></div><br><div class="gmail_quote"><div dir="ltr">On Fri, Sep 28, 2018 at 12:42 PM Jed Brown <<a href="mailto:jed@jedbrown.org">jed@jedbrown.org</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Michael Werner <<a href="mailto:michael.werner@dlr.de" target="_blank">michael.werner@dlr.de</a>> writes:<br>
<br>
> A low-Mach preconditioner might help with the current test cases, <br>
> however I also intend to apply this code to high-Mach number <br>
> flows, so I need to find a more general solution. Actually, the <br>
> high-Mach number applications are more important, since so far all <br>
> the low-Mach cases are small enough to be solved with direct <br>
> solvers.<br>
><br>
> I was also thinking about using a geometric multigrid approach via <br>
> DMPlex. As far as I understood, hyperbolic problems are difficult <br>
> to solve with AMG because the solver isn't aware of the underlying <br>
> structure of the problem. Therefore I would think that a geometric <br>
> multigrid approach should produce better results, right? Do you <br>
> think it would be worthwile to implement a DMPlex, or would I <br>
> still run into the same problems?<br>
<br>
Agglomeration-based geometric multigrid with block relaxation is a<br>
common technique for unstructured compressible CFD, but there are a lot<br>
of subtleties. See FUN3D papers (Diskin, Thomas, sometimes Brandt) for<br>
examples of this approach.<br>
</blockquote></div>