<div>Thanks Tom,</div>
<div> </div>
<div>And I found CCTK_Proc#.out files were created. I emailed you before I had a very strange problem that I couldn't use all the cpu cores for some parameter files. I found that the problem seems to be caused by httpd thorn but cure in Cactus development version. I don't know the reason. Anyway thanks for your quick reply.
</div>
<div> </div>
<div>Regards,</div>
<div> </div>
<div>Hee Il <br><br> </div>
<div><span class="gmail_quote">2007/2/12, Tom Goodale <<a href="mailto:goodale@cct.lsu.edu">goodale@cct.lsu.edu</a>>:</span>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">Hi,<br><br>you only need ghost-zones on internal processor boundaries. So with 9<br>procs you get 2x2x2 procs with 32 physical points each and one ghost zone
<br>each.<br><br>Cheers,<br><br>Tom<br><br><br>On Mon, 12 Feb 2007, Hee Il Kim wrote:<br><br>> Hi,<br>><br>> I have several questions about pugh. When I run a 64 cubed single<br>> cpu simulation, the cactus run shows
<br>><br>> INFO (CartGrid3D): x=>[0,63] y=>[0,63] z=>[0,63]<br>> INFO (PUGH): MPI Evolution on 1 processors<br>> INFO (PUGH): 3-dimensional grid functions<br>> INFO (PUGH): Size: 64 64 64<br>> INFO (PUGH): Processor topology: 1 x 1 x 1
<br>> INFO (PUGH): Local load: 262144 [64 x 64 x 64]<br>><br>> Nothing strange. But when I make a mpirun, for example, using 8 cpus (=8 cpu<br>> cores),<br>><br>> INFO (CartGrid3D): x=>[0,63] y=>[0,63] z=>[0,63]
<br>> INFO (PUGH): MPI Evolution on 8 processors<br>> INFO (PUGH): 3-dimensional grid functions<br>> INFO (PUGH): Size: 64 64 64<br>> INFO (PUGH): Processor topology: 2 x 2 x 2<br>> INFO (PUGH): Local load: 35937 [33 x 33 x 33]
<br>><br>> Why is the local load 33x33x33? I think it should be 32x32x32 or 34x34x34<br>> if it includes ghostzones with stencil 1 ? I can get a correct result even<br>> with this problem. What does the local load represent?
<br>><br>> In order to look into further, I tried to use CCTK_MyProcs function. But I<br>> only get rank 0. In fact I don't know how to display the rank of computing<br>> node cpus. I tried "-r[o|e]" option for cactus run but no lucks.
<br>><br>> I'm using 6 node intel dual core (presler 930) cluster. I use Cactus for<br>> numerical relativity but I'm a newbie on computational science. I appreciate<br>> your kind helps in advance.<br>
><br>> Kim, Hee Il<br>> from Korea<br>> <a href="mailto:heeilkim@gmail.com">heeilkim@gmail.com</a><br>><br>_______________________________________________<br>Users mailing list<br><a href="mailto:Users@cactuscode.org">
Users@cactuscode.org</a><br><a href="http://www.cactuscode.org/mailman/listinfo/users">http://www.cactuscode.org/mailman/listinfo/users</a><br></blockquote></div><br>