<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Sun, Feb 5, 2017 at 9:40 PM, Barry Smith <span dir="ltr"><<a href="mailto:bsmith@mcs.anl.gov" target="_blank">bsmith@mcs.anl.gov</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><br>
Is there any way we can substitute a deterministic random number generators for the bullshit partitioning packages that insist on using "system" non-portable random number generators? We should have tests that use metis/parmetis and chaco or whatever but we shouldn't have to deal with their bad decisions. Matt, what horrible partitioning is being used here?<br></blockquote><div><br></div><div>I put in something that accepts fixed partitions, and it should be used for all parallel tests. I converted a bunch of them already, but its</div><div>possible I missed this one.</div><div><br></div><div>The workflow for conversion is</div><div><br></div><div> a) Run the test with the -partition_view flag</div><div><br></div><div> b) The partition of cells prints out at the top</div><div><br></div><div> c) Copy that partition into a struct with the number of cells in one array and the cells numbers in another. You can see an example in Plex tests ex1.</div><div><br></div><div> d) Create a Shell Partitioner. Again refer to ex1</div><div><br></div><div> Matt</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Barry<br>
<br>
not ok diff-snes_tutorials-ex12_tri_<wbr>bddc<br>
# 2,12c2,11<br>
# < 0 KSP Residual norm 11.5911<br>
# < 1 KSP Residual norm 0.419162<br>
# < 2 KSP Residual norm 0.0466633<br>
# < 3 KSP Residual norm 0.00282513<br>
# < 4 KSP Residual norm 0.000140711<br>
# < 5 KSP Residual norm 2.35332e-05<br>
# < 6 KSP Residual norm 8.28552e-07<br>
# < 7 KSP Residual norm 1.43643e-08<br>
# < 8 KSP Residual norm 2.827e-10<br>
# < Linear solve converged due to CONVERGED_RTOL iterations 8<br>
# < 1 SNES Function norm 7.836e-10<br>
# ---<br>
# > 0 KSP Residual norm 11.6648<br>
# > 1 KSP Residual norm 0.443991<br>
# > 2 KSP Residual norm 0.0393777<br>
# > 3 KSP Residual norm 0.00187772<br>
# > 4 KSP Residual norm 0.000140435<br>
# > 5 KSP Residual norm 1.08843e-05<br>
# > 6 KSP Residual norm 2.87537e-07<br>
# > 7 KSP Residual norm 5.20043e-09<br>
# > Linear solve converged due to CONVERGED_RTOL iterations 7<br>
# > 1 SNES Function norm 1.59735e-08<br>
# 18c17<br>
# < total number of linear solver iterations=8<br>
# ---<br>
# > total number of linear solver iterations=7<br>
# 76,77c75,76<br>
# < BDDC: Global dofs sizes: all 225 interface 43 coarse 7<br>
# < BDDC: Coarsening ratios: all/coarse 32 interface/coarse 6<br>
# ---<br>
# > BDDC: Global dofs sizes: all 225 interface 57 coarse 9<br>
# > BDDC: Coarsening ratios: all/coarse 25 interface/coarse 6<br>
# 81,84c80,83<br>
# < BDDC: Interior dofs : 21 49 36<br>
# < BDDC: Interface dofs : 15 22 17<br>
# < BDDC: Primal dofs : 3 4 3<br>
# < BDDC: Local dofs : 36 64 54<br>
# ---<br>
# > BDDC: Interior dofs : 21 42 33<br>
# > BDDC: Interface dofs : 15 36 23<br>
# > BDDC: Primal dofs : 3 6 4<br>
# > BDDC: Local dofs : 36 78 57<br>
# 99,100c98,99<br>
# < rows=7, cols=7<br>
# < total: nonzeros=35, allocated nonzeros=43<br>
# ---<br>
# > rows=9, cols=9<br>
# > total: nonzeros=61, allocated nonzeros=68<br>
# 115c114<br>
# < factor fill ratio given 5., needed 1.17143<br>
# ---<br>
# > factor fill ratio given 5., needed 1.13115<br>
# 119c118<br>
# < rows=7, cols=7<br>
# ---<br>
# > rows=9, cols=9<br>
# 121c120<br>
# < total: nonzeros=41, allocated nonzeros=41<br>
# ---<br>
# > total: nonzeros=69, allocated nonzeros=69<br>
# 127,128c126,127<br>
# < rows=7, cols=7<br>
# < total: nonzeros=35, allocated nonzeros=35<br>
# ---<br>
# > rows=9, cols=9<br>
# > total: nonzeros=61, allocated nonzeros=61<br>
# 130c129<br>
# < not using I-node routines<br>
# ---<br>
# > using I-node routines: found 7 nodes, limit used is 5<br>
# 135c134<br>
# < total: nonzeros=1585, allocated nonzeros=1764<br>
# ---<br>
# > total: nonzeros=1625, allocated nonzeros=1855<br>
<br>
</blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature" data-smartmail="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>