<div dir="ltr"><p class="MsoNormal" style="margin:0cm 0cm 8pt;line-height:107%"><font size="4" style="" face="monospace"><a name="_GoBack"></a><span lang="EN-US" style="">Hello all,</span></font></p>
<p class="MsoNormal" style="margin:0cm 0cm 8pt;line-height:107%"><font size="4" face="monospace"><span lang="EN-US">It souds that the best way to introduce
petsc in a code is not to introduce it, but develop the code over the petsc
structure.</span></font></p><p class="MsoNormal" style="margin:0cm 0cm 8pt;line-height:107%"><font face="monospace"><span lang="EN-US" style="font-size:large">It is probably true but my problem is that my existing code </span><span lang="EN-US" style=""><font size="4">already is equipped with a</font></span><font size="4"> domain decomposition based on MPI (</font><span lang="EN-US" style="font-size:large">a typical themal hydraulic with cartesian
staggered mesh)</span></font></p><p class="MsoNormal" style="margin:0cm 0cm 8pt;line-height:107%"><font size="4" face="monospace">The user can slice the domain in
sub-domains and construct a linear problem by block: each sub-domain assembles
its part of the operator and its part of the RHS.</font></p><p class="MsoNormal" style="margin:0cm 0cm 8pt;line-height:107%"><span style="font-size:large"><font face="monospace">I am wondering what is the best way now to introduce petsc (considering that I don’t want to assemble a global
operator on a given proc). Is there an example that would show how to
introduce petsc in this situation ?</font></span></p><div><font face="monospace" size="4">Thank you, </font><span style="font-size:large;font-family:monospace">Pierre </span><span style="font-size:large;font-family:monospace"> </span><span style="font-size:large;font-family:monospace"> </span></div></div>