On Thu, Dec 23, 2010 at 7:46 AM, Mark F. Adams <span dir="ltr"><<a href="mailto:adams@pppl.gov">adams@pppl.gov</a>></span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
What other games are there for scaleable partitioners?<br>
<br>
Scotch looks interesting from their web page, does anyone have any experience with it? They seem to claim 2K processor runs which is about one or two orders short of showing that it might scale.<br>
<br>
N.B. I've tried Chaco but found it was not even parallel.</blockquote><div><br></div><div>Also, I do not know if any other package does multiobjective partitioning like ParMetis does.</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;"><font color="#888888"><br>
Mark</font><div><div></div><div class="h5"><br>
<br>
On Dec 23, 2010, at 8:44 AM, Barry Smith wrote:<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
On Dec 22, 2010, at 9:58 PM, Mark F. Adams wrote:<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
I've been concerned for a while that, as Barry says, ParMetis seems orphaned. I've had problems with scaling ParMetis up on BG for a while. I believe they allocate a PxP (integer) matrix which kills my code (my code is flat MPI but I make sub communicators so that each ParMetis MPI process has the whole node to itself). Can anyone scale Parmetis up on BG-P?<br>
<br>
Anyway this has been on my wish list for a while so I thought I'd use this opportunity to see if anyone has ideas for a solution.<br>
</blockquote>
<br>
Switch to a non-orphaned partitioning package. Just like Microsoft Windows isn't the only game in town, ParMetis isn't the only game in town.<br>
<br>
Barry<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
Mark<br>
<br>
On Dec 22, 2010, at 7:50 PM, Matthew Knepley wrote:<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
On Wed, Dec 22, 2010 at 3:02 PM, Lisandro Dalcin <<a href="mailto:dalcinl@gmail.com" target="_blank">dalcinl@gmail.com</a>> wrote:<br>
On 22 December 2010 19:57, Barry Smith <<a href="mailto:bsmith@mcs.anl.gov" target="_blank">bsmith@mcs.anl.gov</a>> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
On Dec 22, 2010, at 4:51 PM, Lisandro Dalcin wrote:<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
On 21 December 2010 20:50, Sean Farley <<a href="mailto:sean@mcs.anl.gov" target="_blank">sean@mcs.anl.gov</a>> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
fixing/understanding 64-bit checks,<br>
</blockquote>
<br>
I think these checks are ok, please report problems.<br>
</blockquote>
<br>
Would it be possible to get ParMetis working with 64bit indices?<br>
Sean<br>
</blockquote>
<br>
This is from parmetis.h<br>
<br>
/* Indexes are as long as integers for now */<br>
#ifdef IDXTYPE_INT<br>
typedef int idxtype;<br>
#else<br>
typedef short idxtype;<br>
#endif<br>
<br>
Have any of you seriously tryied parmetis in 64 bits?<br>
</blockquote>
<br>
Yes, when I noticed that block of code several years ago. Looks like it would be easy to switch to long int but sadly this include is misleading about the simplicity, turns out this typedef beasty is not used throughout the source code consistently.<br>
</blockquote>
<br>
Damn,<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
I sent email about the 64 bit int at that time and got back the response "in a few months". Note that his non-MPI codes may fully support the 64 bit (I'm guessing that is where the $ is so is all he cares about). That was like two years ago. I consider Parmetis orphaned, not maintained, hence in the HPC world, dead.<br>
<br>
</blockquote>
<br>
Add to that license issues. A fork in order to maintain and improve it<br>
is unlikely.<br>
<br>
If this is a big deal, I can mail George directly. He let me fool with the tarball and distribute it<br>
with PETSc last time I asked him.<br>
<br>
Matt<br>
<br>
--<br>
Lisandro Dalcin<br>
---------------<br>
CIMEC (INTEC/CONICET-UNL)<br>
Predio CONICET-Santa Fe<br>
Colectora RN 168 Km 472, Paraje El Pozo<br>
Tel: +54-342-4511594 (ext 1011)<br>
Tel/Fax: +54-342-4511169<br>
<br>
<br>
<br>
-- <br>
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<br>
</blockquote>
<br>
</blockquote>
<br>
<br>
</blockquote>
<br>
</div></div></blockquote></div><br><br clear="all"><br>-- <br>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<br>