<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Mon, May 4, 2015 at 5:57 PM, Reza Yaghmaie <span dir="ltr"><<a href="mailto:reza.yaghmaie2@gmail.com" target="_blank">reza.yaghmaie2@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div><br></div><div>Dear Matt,</div><div><br></div><div>The initial guess was zero for all cases of SNES solvers. The initial jacobian was identity for all cases. The system is small and is ran sequentially.</div><div>I have to add that I use FDColoring routine for the jacobian as well.</div></blockquote><div><br></div><div>So, the initial Jacobian was the identity, but we have no idea how close that is to the true Jacobian. You could use the</div><div>coloring routines (the default) combined with</div><div><br></div><div> <a href="http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/SNES/SNESQNSetScaleType.html">http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/SNES/SNESQNSetScaleType.html</a></div><div><br></div><div>using Jacobian, which sets the initial guess to the true Jacobian. That would give us an idea whether the initial guess</div><div>impacted QN covergence.</div><div><br></div><div> Thanks,</div><div><br></div><div> Matt</div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div>Regards,</div><div>Ray</div><div><br></div><div><br></div><div><br></div><br><br>On Monday, May 4, 2015, Matthew Knepley <<a href="mailto:knepley@gmail.com" target="_blank">knepley@gmail.com</a>> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Mon, May 4, 2015 at 5:41 PM, Reza Yaghmaie <span dir="ltr"><<a>reza.yaghmaie2@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div><br></div><div>Dear Matt,</div><div><br></div><div>Actually the initial jacobian was identity. Regular SNES converges in 48 iterations, GMRES in 19, NCG in 67,...</div><div>Do you think SNESQN with the basiclineseach was the problem for divergence?</div><div>If I use SNESQN by default should not it converge with initial identity jacobian?</div></blockquote><div><br></div><div>Do you mean that you used an initial guess of the identity, or that the Actual Jacobian was the identity at your</div><div>initial guess?</div><div><br></div><div> Matt</div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div>Best regards,</div><div>Reza</div><div><br></div><div><br></div><br><br>On Monday, May 4, 2015, Matthew Knepley <<a>knepley@gmail.com</a>> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Mon, May 4, 2015 at 2:11 PM, Reza Yaghmaie <span dir="ltr"><<a>reza.yaghmaie2@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div dir="ltr"><br><div>Dear PETSC representatives,</div><div><br></div><div>I am solving a nonlinear problem with SNESNGMRES and it converges faster with less iterations compared to otehr SNES methods. Any idea why that is the case?</div></div></blockquote><div><br></div><div>It is impossible to tell with this information.</div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div dir="ltr"><div>Also SNESQN diverges quickly. I tried to use <span style="color:rgb(0,0,0);white-space:pre-wrap">SNESLINESEARCHBASIC for the linesearch option and nothing changes. </span></div></div></blockquote><div><br></div><div>This can happen, especially if your matrix is far from the identity.</div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div dir="ltr"><div><font color="#000000"><span style="white-space:pre-wrap">I presume it is using the default SNESQN method. Btw, there are three options for QN, as "</span></font>SNES_QN_LBFGS, SNES_QN_BROYDEN, SNES_QN_BADBROYEN<span style="white-space:pre-wrap;color:rgb(0,0,0)">" in teh manual. I tried to associate them with SNES however it seems these hyphened names don't work there. What am I missing?</span></div></div></blockquote><div><br></div><div>-snes_qn_scale_type <lbfgs,broyden,badbroyden> <br></div><div><br></div><div>from <a href="http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/SNES/SNESQNSetType.html" target="_blank">http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/SNES/SNESQNSetType.html</a></div><div><br></div><div> Thanks,</div><div><br></div><div> Matt</div><div><br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div dir="ltr"><div><span style="white-space:pre-wrap;color:rgb(0,0,0)">Best regards,</span></div><div><font color="#000000"><span style="white-space:pre-wrap">Ray</span></font></div></div>
</blockquote></div><br><br clear="all"><span class=""><font color="#888888"><span><font color="#888888"><div><br></div>-- <br><div>What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.<br>-- Norbert Wiener</div>
</font></span></font></span></div></div><span class=""><font color="#888888">
</font></span></blockquote><span class=""><font color="#888888">
</font></span></blockquote></div><span class=""><font color="#888888"><br><br clear="all"><div><br></div>-- <br><div>What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.<br>-- Norbert Wiener</div>
</font></span></div></div>
</blockquote>
</blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature">What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.<br>-- Norbert Wiener</div>
</div></div>