<div dir="ltr">Hi Changqing,<div><br></div><div>this <a href="https://urldefense.us/v3/__https://petsc.org/release/manual/dmplex/*metric-based-mesh-adaptation__;Iw!!G_uCfscf7eWS!eB6Mm6bN2Re49sXqbWzUKjZYrgjZPTW4yai322b1f2lqFxA0sTnZ79gC9LA-eJaZQzkLNUMW0r8O9R7Ng0lsobmGo-80$">Metric Based Adaptation</a> might be helpful for you</div><div><br></div><div>Thanks</div><div>Ravindra<br><br></div></div><br><div class="gmail_quote gmail_quote_container"><div dir="ltr" class="gmail_attr">On Wed, Jan 15, 2025 at 10:21 PM Ye Changqing <<a href="mailto:Ye_Changqing@outlook.com">Ye_Changqing@outlook.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Dear PETSc developers,<br>
<br>
I am curious about the mystery behind ``refine'' and ``coarsen'' routines of DM. <br>
<br>
- As mentioned on the webpage, these routines appear to neglect the coordinate information associated with the current DM. I suspect that some graph-based algebraic operations might be involved. Could you kindly share the underlying design principles or implementation strategies behind these routines?<br>
<br>
Best,<br>
Changqing</blockquote></div>