<div dir="ltr"><div class="gmail_default" style="font-family:arial,helvetica,sans-serif;font-size:small">Thanks a lot.</div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif;font-size:small">I'm going to change to the interlaced version. It is easier and more efficient.</div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif;font-size:small">Best regards.</div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif;font-size:small">Bernardo </div></div><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Feb 15, 2017 at 12:07 AM, 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"><span class=""><br>
> On Feb 14, 2017, at 7:57 PM, Bernardo Rocha <<a href="mailto:bernardomartinsrocha@gmail.com">bernardomartinsrocha@gmail.<wbr>com</a>> wrote:<br>
><br>
> No particular reason. Let's say it was the first approach I tried in an old code.<br>
><br>
> OK, so to get it working I should use the interlaced version for the ordering of my dofs.<br>
> It does not require too much coding on my side.<br>
><br>
> Just to know, if I wanted to insist in the non-interlaced, to keep using Hypre/BoomerAMG from PETSc I would need to write some code in PETSc for wrapping this feature (AMGSetDofFunc), right?<br>
><br>
> Thanks.<br>
><br>
> On Tue, Feb 14, 2017 at 11:48 PM, Barry Smith <<a href="mailto:bsmith@mcs.anl.gov">bsmith@mcs.anl.gov</a>> wrote:<br>
><br>
> Ahh, thanks for the explanation. What a nutso API: way to general, they should just support interlaced and non-interlaced.<br>
><br>
> Anyways we strongly recommend using the interlaced version; for most calculations it gives much better locality for cache lines and cache. We always use interlaced. Any particular reason you use non-interlaced?<br>
<br>
</span> Yes, you would have to add a call to HYPRE_BoomerAMGSetDofFunc() with the information as you computed. For trying things out the easily way would be to just stick it in an appropriate place in hypre.c<br>
<span class="HOEnZb"><font color="#888888"><br>
Barry<br>
</font></span><div class="HOEnZb"><div class="h5"><br>
<br>
<br>
<br>
><br>
> Barry<br>
><br>
> > On Feb 14, 2017, at 7:17 PM, Bernardo Rocha <<a href="mailto:bernardomartinsrocha@gmail.com">bernardomartinsrocha@gmail.<wbr>com</a>> wrote:<br>
> ><br>
> > Thanks a lot for the reply.<br>
> ><br>
> > From my understanding and specifically for my problem I would have to use the AMGSetDofFunc as follows:<br>
> ><br>
> > nsyseq = 3;<br>
> > for (i=0; i<nsyseq; i++)<br>
> > for (k=0; k<N; k++) // N: number of nodes<br>
> > {<br>
> > eqnIndex[i*N+k] = i;<br>
> > }<br>
> > HYPRE_BoomerAMGSetDofFunc(<wbr>preconditioner, eqnIndex);<br>
> ><br>
> > For a system of 3 PDEs (e.g. linear elasticity in 3D) the default behaviour of the nodal coarsening in BoomerAMG, considers an array that is filled with<br>
> ><br>
> > eqnIndex = 0 1 2 0 1 2 0 1 2....0 1 2<br>
> ><br>
> > > HYPRE_BoomerAMGSetDofFunc<br>
> ><br>
> > I could not find an indication of that this is suppose to be setting; it is mentioned in the users manual but I don't know what it means. It looks like it takes an integer array but I don't even know how long that array is.<br>
> ><br>
> > If you figure out what it means and can add support for it we'll take it as a pull request.<br>
> ><br>
> > OK Barry, I will try to specify the block size of the matrix.<br>
> > But I'm not sure if the block size would be 3 or N (number of nodes).<br>
> > This is my ordering of the displacements: [ux1, ux2, ..., uxN, uy1, uy2,...uyN, uz1, uz2, ..., uzN]<br>
> ><br>
> > > HYPRE_BoomerAMGSetNumFunctions<br>
> ><br>
> > Set the block size of the matrix; this information is then transferred automatically to hypre. Note that you can set a block size even for AIJ matrices.<br>
> > <br>
> > Best regards,<br>
> > Bernardo<br>
><br>
><br>
<br>
</div></div></blockquote></div><br></div>