From nek5000-users at lists.mcs.anl.gov Mon Nov 2 10:39:06 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 2 Nov 2015 19:39:06 +0300 Subject: [Nek5000-users] energy dissipation over a subdomain In-Reply-To: References: Message-ID: Hi, Alex! Thank you very much, sorry for delay, I've been travelling. Best, Ilias On Fri, Oct 30, 2015 at 6:21 PM, wrote: > Hi Ilias, > > As I mentioned to you earlier, for the case when you subdomain integration > consists of whole elements, I would do the simplest > > dimension ybm1(lx1*ly1*lz1), rmask(lx1*ly1*lz1,lelt) > > nxyz = nx1*ny1*nz1 > ntot = nxzy*nelt > do e=1,nelt > call rzero(rmask,ntot) > call col3(ybm1,bm1,nxyz) > yavg = vlsum(ybm1,nxyz) > yavg = yavg/vlsum(ym1,nxyz) > if (yavg.gt.y_something) call rone(rmask(1,1,1,e),nxyz) > enddo > > and then use glsc3 with arguments of bm1, your quantity on GLL mesh, and > created rmask. > > Note that the above approach with modification will work even for the case > of the integration subdomain boundary cutting through the elements but you > need to use every GLL's ym1 instead of yavg -- be aware that the spectral > convergence of this integration with increase of polynomial order will be > affected since it is equivalent to integration of a step/discontinuous > function inside an element... > > Note also that if the integration subdomain is much smaller than the whole > domain, you could compute the collocation with your quantity and bm1 inside > the above loop and do summation yourself > > Aleks > > > ------------------------------ > *From:* nek5000-users-bounces at lists.mcs.anl.gov [ > nek5000-users-bounces at lists.mcs.anl.gov] on behalf of > nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] > *Sent:* Thursday, October 29, 2015 1:16 PM > *To:* nek5000-users at lists.mcs.anl.gov > *Subject:* [Nek5000-users] energy dissipation over a subdomain > > Dear neks! > > I would like to compute kinetic energy dissipation over different > subdomains. > > It involves the calculation of the dissipation itself in GLL points, > and integration over some domain. > > I understand now that it is more accurate to involve whole number of > elements > for integration, > and that for computation of the integral over > a subdomain one can use a mask (of 1s and 0es) over the whole domain, > and after that to use global scalar product like glsc3. > > Are there the implementations of these ideas? > first of all, the creation of a mask, based on geometrical constructions > like > IF (Y.gt.something) THEN > ... > > and also if somebody has already a procedure of kinetic energy dissipation, > though here it is more strait: calculation of of velocity gradient and > contraction of > its multiplication with itself, > > Thank you, Ilias > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Tue Nov 3 13:46:03 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 3 Nov 2015 20:46:03 +0100 Subject: [Nek5000-users] Where to modify the equations solved in Nek5000? Message-ID: Dear all, I am completely new to Nek5000. I would like to use Nek5000 for the simulations of complex fluids like non-Newtonian fluids. Where should I look into for the equations? There seems no information on this in the tutorial. Thanks a lot. Mengqi -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Tue Nov 3 15:27:36 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 3 Nov 2015 16:27:36 -0500 Subject: [Nek5000-users] new to nek5000 In-Reply-To: References: Message-ID: hey KS i followed all the steps. i am having issue with step 3. How do i change the path of the mpi compiler? Thanx AVR Apoorva Vinayak Rudra Graduate Research Assistant Department of Chemical Engineering City College of New York (CCNY) On Fri, Oct 30, 2015 at 2:32 PM, wrote: > Hi AVR, > > Here is a minimal info for running a first case: > > ww.mcs.anl.gov/~obabko/nekrun.pdf > > but I would recommend to look at the available literature: > > http://nek5000.mcs.anl.gov/documentation-2/ > > Aleks > > ------------------------------ > *From:* nek5000-users-bounces at lists.mcs.anl.gov [ > nek5000-users-bounces at lists.mcs.anl.gov] on behalf of > nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] > *Sent:* Monday, October 26, 2015 3:12 PM > *To:* nek5000-users at lists.mcs.anl.gov > *Subject:* [Nek5000-users] new to nek5000 > > hey I am extremely new to nek5000 > I am trying to solve the examples that are already present. > If I just want to compile the already existing files, what are the steps? > This may seem lame but cant help it..sry > AVR > > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Tue Nov 3 15:58:06 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 3 Nov 2015 16:58:06 -0500 Subject: [Nek5000-users] new to nek5000 In-Reply-To: References: Message-ID: hey Aleks can you send me the complete file from this link you posted: ww.mcs.anl.gov/~obabko/nekrun.pdf thx AVR Apoorva Vinayak Rudra Graduate Research Assistant Department of Chemical Engineering City College of New York (CCNY) On Tue, Nov 3, 2015 at 4:27 PM, Apoorva Rudra wrote: > hey KS > i followed all the steps. i am having issue with step 3. How do i change > the path of the mpi compiler? > Thanx > > AVR > > Apoorva Vinayak Rudra > Graduate Research Assistant > Department of Chemical Engineering > City College of New York (CCNY) > > > On Fri, Oct 30, 2015 at 2:32 PM, wrote: > >> Hi AVR, >> >> Here is a minimal info for running a first case: >> >> ww.mcs.anl.gov/~obabko/nekrun.pdf >> >> but I would recommend to look at the available literature: >> >> http://nek5000.mcs.anl.gov/documentation-2/ >> >> Aleks >> >> ------------------------------ >> *From:* nek5000-users-bounces at lists.mcs.anl.gov [ >> nek5000-users-bounces at lists.mcs.anl.gov] on behalf of >> nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] >> *Sent:* Monday, October 26, 2015 3:12 PM >> *To:* nek5000-users at lists.mcs.anl.gov >> *Subject:* [Nek5000-users] new to nek5000 >> >> hey I am extremely new to nek5000 >> I am trying to solve the examples that are already present. >> If I just want to compile the already existing files, what are the steps? >> This may seem lame but cant help it..sry >> AVR >> >> >> _______________________________________________ >> Nek5000-users mailing list >> Nek5000-users at lists.mcs.anl.gov >> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users >> >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu Nov 5 11:43:21 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 5 Nov 2015 18:43:21 +0100 Subject: [Nek5000-users] website of NEK5000 In-Reply-To: References: Message-ID: Thank you, Oana, But I still miss a page with the (human readable) brief description of such funcions as avg_all, glsc*, opcopy, comp_vort and so on ... Of course there is the source code in doxygen, but there are no comments. So that description on the old site was really handy for me. Best regards, Ilias On Thu, Oct 29, 2015 at 11:41 PM, wrote: > Hi Ilias, > > Most of that information has been moved to the new manual or the math > functions are in doxygen. We are working now on updating and making that > documentation browasable and link to the new pages properly. > > Sorry for the inconvenience, and please check once again next Monday when > I think we'll be done. > > Oana > ------------------------------ > *From:* nek5000-users-bounces at lists.mcs.anl.gov [ > nek5000-users-bounces at lists.mcs.anl.gov] on behalf of > nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] > *Sent:* Thursday, October 29, 2015 11:50 AM > *To:* nek5000-users at lists.mcs.anl.gov > *Subject:* [Nek5000-users] website of NEK5000 > > Dear neks, > The old version of the site had some pages which are absent in the new > version, > f.e. genbox description, boundary definitions, > and especially description of useful functions. > Is it possible to give a link to the old version of the site, > since the new information does not contain all the information? > > Thank you, Ilias > > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu Nov 5 12:05:40 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 5 Nov 2015 18:05:40 +0000 Subject: [Nek5000-users] website of NEK5000 In-Reply-To: References: , Message-ID: Hi Ilias, I see. We were postponing this since we hoping to add more and extend it. I try to update it later today with was available at the time. Oana ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Thursday, November 05, 2015 11:43 AM To: nek5000-users at lists.mcs.anl.gov Subject: Re: [Nek5000-users] website of NEK5000 Thank you, Oana, But I still miss a page with the (human readable) brief description of such funcions as avg_all, glsc*, opcopy, comp_vort and so on ... Of course there is the source code in doxygen, but there are no comments. So that description on the old site was really handy for me. Best regards, Ilias On Thu, Oct 29, 2015 at 11:41 PM, > wrote: Hi Ilias, Most of that information has been moved to the new manual or the math functions are in doxygen. We are working now on updating and making that documentation browasable and link to the new pages properly. Sorry for the inconvenience, and please check once again next Monday when I think we'll be done. Oana ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Thursday, October 29, 2015 11:50 AM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] website of NEK5000 Dear neks, The old version of the site had some pages which are absent in the new version, f.e. genbox description, boundary definitions, and especially description of useful functions. Is it possible to give a link to the old version of the site, since the new information does not contain all the information? Thank you, Ilias _______________________________________________ Nek5000-users mailing list Nek5000-users at lists.mcs.anl.gov https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Fri Nov 6 13:45:24 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 6 Nov 2015 11:45:24 -0800 Subject: [Nek5000-users] Newest version of svn repo (1088) does not compile Message-ID: Just an FYI for those of you who tend to do svn updates fairly often for nek5000: The svn version 1088 does not compile correctly. I found if I reverted to version 1078 it did compile correctly. Janet -- Janet Scheel Associate Professor of Physics Occidental College 1600 Campus Road, M21 Los Angeles, CA 90041 (323) 259-2777 jscheel at oxy.edu -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Fri Nov 6 14:28:45 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 6 Nov 2015 20:28:45 +0000 Subject: [Nek5000-users] Newest version of svn repo (1088) does not compile In-Reply-To: References: Message-ID: Hi Janet, My fault... we're working on it on this end. Did you update your "makenek" from the repo? Paul ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Friday, November 06, 2015 1:45 PM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Newest version of svn repo (1088) does not compile Just an FYI for those of you who tend to do svn updates fairly often for nek5000: The svn version 1088 does not compile correctly. I found if I reverted to version 1078 it did compile correctly. Janet -- Janet Scheel Associate Professor of Physics Occidental College 1600 Campus Road, M21 Los Angeles, CA 90041 (323) 259-2777 jscheel at oxy.edu -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Sun Nov 8 12:06:07 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sun, 8 Nov 2015 19:06:07 +0100 Subject: [Nek5000-users] Newest version of svn repo (1088) does not compile In-Reply-To: References: Message-ID: Hi Nek, i hope someone can help me to understand. I simulate a ribbed surface (triangulare riblets). When i use genmap command i have the following error: Input mesh tolerance (default 0.2): NOTE: smaller is better, but generous is more forgiving for bad meshes. 0.0001 reading .rea file data ... start locglob_lexico: 8 1428840 11430720 1.00000000000000005E-004 locglob: 1 1 11430720 locglob: 2 64 11430720 locglob: 3 14720 11430720 locglob: 1 942080 11430720 locglob: 2 942080 11430720 locglob: 3 942080 11430720 locglob: 1 942080 11430720 locglob: 2 942080 11430720 locglob: 3 942080 11430720 717319 2 4 Matrix: SELF!! 1 SELF!! 14465 14465 14466 14466 2 SELF!! 29185 29185 29186 29186 cont: SELF!! 717319 ?? ABORT: SELF-CHK 1 3 717319 0 Try to tighten the mesh tolerance! 0 quit STOP Obviously i tried more different values for mesh tolerance. So my question is: there is or there are particular conditions to choose this value? Or depending in which criterion i could choose this value? On 06/11/2015 21:28, nek5000-users at lists.mcs.anl.gov wrote: > > Hi Janet, > > My fault... we're working on it on this end. > > Did you update your "makenek" from the repo? > > Paul > > ------------------------------------------------------------------------ > *From:* nek5000-users-bounces at lists.mcs.anl.gov > [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of > nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] > *Sent:* Friday, November 06, 2015 1:45 PM > *To:* nek5000-users at lists.mcs.anl.gov > *Subject:* [Nek5000-users] Newest version of svn repo (1088) does not > compile > > Just an FYI for those of you who tend to do svn updates fairly often > for nek5000: > > The svn version 1088 does not compile correctly. I found if I reverted > to version 1078 it did compile correctly. > > Janet > > -- > Janet Scheel > Associate Professor of Physics > Occidental College > 1600 Campus Road, M21 > Los Angeles, CA 90041 > (323) 259-2777 > jscheel at oxy.edu > > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Sun Nov 8 14:24:39 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sun, 8 Nov 2015 20:24:39 +0000 Subject: [Nek5000-users] Newest version of svn repo (1088) does not compile In-Reply-To: References: , Message-ID: This could be due several different problems... A likely candidate is if your mesh has degenerate elements that have two vertices collapsed together. I would suggest inspecting the element number in question as a starting point. Other options include working with a subset of your mesh to try to uncover the issue. You can look at the genmap source (tools/genmap/genmap.f) to see what is being printed out. It appears that genmap doesn't like element number 717319 Paul ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Sunday, November 08, 2015 12:06 PM To: nek5000-users at lists.mcs.anl.gov Subject: Re: [Nek5000-users] Newest version of svn repo (1088) does not compile Hi Nek, i hope someone can help me to understand. I simulate a ribbed surface (triangulare riblets). When i use genmap command i have the following error: Input mesh tolerance (default 0.2): NOTE: smaller is better, but generous is more forgiving for bad meshes. 0.0001 reading .rea file data ... start locglob_lexico: 8 1428840 11430720 1.00000000000000005E-004 locglob: 1 1 11430720 locglob: 2 64 11430720 locglob: 3 14720 11430720 locglob: 1 942080 11430720 locglob: 2 942080 11430720 locglob: 3 942080 11430720 locglob: 1 942080 11430720 locglob: 2 942080 11430720 locglob: 3 942080 11430720 717319 2 4 Matrix: SELF!! 1 SELF!! 14465 14465 14466 14466 2 SELF!! 29185 29185 29186 29186 cont: SELF!! 717319 ?? ABORT: SELF-CHK 1 3 717319 0 Try to tighten the mesh tolerance! 0 quit STOP Obviously i tried more different values for mesh tolerance. So my question is: there is or there are particular conditions to choose this value? Or depending in which criterion i could choose this value? On 06/11/2015 21:28, nek5000-users at lists.mcs.anl.gov wrote: Hi Janet, My fault... we're working on it on this end. Did you update your "makenek" from the repo? Paul ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Friday, November 06, 2015 1:45 PM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Newest version of svn repo (1088) does not compile Just an FYI for those of you who tend to do svn updates fairly often for nek5000: The svn version 1088 does not compile correctly. I found if I reverted to version 1078 it did compile correctly. Janet -- Janet Scheel Associate Professor of Physics Occidental College 1600 Campus Road, M21 Los Angeles, CA 90041 (323) 259-2777 jscheel at oxy.edu _______________________________________________ Nek5000-users mailing list Nek5000-users at lists.mcs.anl.gov https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Sun Nov 8 16:37:04 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sun, 8 Nov 2015 23:37:04 +0100 Subject: [Nek5000-users] dissipation Message-ID: Dear neks, I am trying to user comp_sij to compute dissipation in every point of the domain. The data storage is organised according to real sij(lx1*ly1*lz1,nij,lelv) so the index of a component of tensor precedes number of element, and the is no continuous block of lx1*ly1*lz1,lelv points for an element of tensor. And so, the global functions like col3 of glsc can not be applied. At least I don't understand how. To bypass this, I copied comp_sij to my usr file and changed the order of last indices in sij real sij(lx1*ly1*lz1,lelv,nij) After that I use the followng code to compute viscous dissipation nij = 3 if (if3d) nij=6 call il_comp_sij(sij,nij,vx,vy,vz,ur,us,ut,vr,vs,vt,wr,ws,wt) call rzero(il_dissip,lv) do j=1,nij call col3(sij2,sij(1,1,1,1,j),sij(1,1,1,1,j),nxyz) il_dissip=il_dissip+sij2 if (if3d .and. j.ge.4) then il_dissip=il_dissip+sij2 end if if ( .not. if3d .and. j.eq.3 ) then il_dissip=il_dissip+sij2 end if il_dissip=2*param(2)*il_dissip enddo Is is a fair way to compute dissipation? Thanks, Ilias -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Sun Nov 8 22:03:04 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 9 Nov 2015 04:03:04 +0000 Subject: [Nek5000-users] dissipation In-Reply-To: References: Message-ID: Hi Ilias, How about: subroutine dissipation(sij2) include 'SIZE' include 'TOTAL' real sij2(lx1*ly1*lz1,lelt) common /scrns/ sij (lx1*ly1*lz1,6,lelv) parameter (lr=lx1*ly1*lz1) common /scruz/ ur(lr),us(lr),ut(lr) $ , vr(lr),vs(lr),vt(lr) $ , wr(lr),ws(lr),wt(lr) integer e nij = 3+3*(ndim-2) call comp_sij(sij,nij,vx,vy,vz,ur,us,ut,vr,vs,vt,wr,ws,wt) nxyz = nx1*ny1*nz1 n = nx1*ny1*nz1*nelv call rzero(sij2,n) do e=1,nelv do j=1,nij do i=1,nxyz sij2(i,e)=sij2(i,e)+sij(i,j,e)**2 enddo enddo enddo return end ? This should function correctly. Best, Paul ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Sunday, November 08, 2015 4:37 PM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] dissipation Dear neks, I am trying to user comp_sij to compute dissipation in every point of the domain. The data storage is organised according to real sij(lx1*ly1*lz1,nij,lelv) so the index of a component of tensor precedes number of element, and the is no continuous block of lx1*ly1*lz1,lelv points for an element of tensor. And so, the global functions like col3 of glsc can not be applied. At least I don't understand how. To bypass this, I copied comp_sij to my usr file and changed the order of last indices in sij real sij(lx1*ly1*lz1,lelv,nij) After that I use the followng code to compute viscous dissipation nij = 3 if (if3d) nij=6 call il_comp_sij(sij,nij,vx,vy,vz,ur,us,ut,vr,vs,vt,wr,ws,wt) call rzero(il_dissip,lv) do j=1,nij call col3(sij2,sij(1,1,1,1,j),sij(1,1,1,1,j),nxyz) il_dissip=il_dissip+sij2 if (if3d .and. j.ge.4) then il_dissip=il_dissip+sij2 end if if ( .not. if3d .and. j.eq.3 ) then il_dissip=il_dissip+sij2 end if il_dissip=2*param(2)*il_dissip enddo Is is a fair way to compute dissipation? Thanks, Ilias -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Mon Nov 9 08:02:13 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 9 Nov 2015 15:02:13 +0100 Subject: [Nek5000-users] dissipation In-Reply-To: References: Message-ID: Thank you, Paul, Only I think to account for all nondiagonal components I have to add conditions to the inner loop like: do e=1,nelv do j=1,nij do i=1,nxyz sij2(i,e)=sij2(i,e)+sij(i,j,e)**2 if (if3d .and. j.ge.4) sij2(i,e)=sij2(i,e)+sij(i,j,e)**2 if ( .not. if3d .and. j.eq.3 ) sij2(i,e)=sij2(i,e)+sij(i,j,e)**2 enddo enddo enddo Best regards, Ilias On Mon, Nov 9, 2015 at 5:03 AM, wrote: > > Hi Ilias, > > > How about: > > subroutine dissipation(sij2) > > > include 'SIZE' > > include 'TOTAL' > > > real sij2(lx1*ly1*lz1,lelt) > > > common /scrns/ sij (lx1*ly1*lz1,6,lelv) > > > parameter (lr=lx1*ly1*lz1) > > common /scruz/ ur(lr),us(lr),ut(lr) > > $ , vr(lr),vs(lr),vt(lr) > > $ , wr(lr),ws(lr),wt(lr) > > > integer e > > > nij = 3+3*(ndim-2) > > > call comp_sij(sij,nij,vx,vy,vz,ur,us,ut,vr,vs,vt,wr,ws,wt) > > > nxyz = nx1*ny1*nz1 > > n = nx1*ny1*nz1*nelv > > > call rzero(sij2,n) > > > do e=1,nelv > > do j=1,nij > > do i=1,nxyz > > sij2(i,e)=sij2(i,e)+sij(i,j,e)**2 > > enddo > > enddo > > enddo > > > return > > end > > ? > > This should function correctly. > > Best, Paul > > > ------------------------------ > *From:* nek5000-users-bounces at lists.mcs.anl.gov [ > nek5000-users-bounces at lists.mcs.anl.gov] on behalf of > nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] > *Sent:* Sunday, November 08, 2015 4:37 PM > *To:* nek5000-users at lists.mcs.anl.gov > *Subject:* [Nek5000-users] dissipation > > Dear neks, > I am trying to user comp_sij to compute dissipation in every point of the > domain. > The data storage is organised according to > real sij(lx1*ly1*lz1,nij,lelv) > so the index of a component of tensor > precedes number of element, and > the is no continuous block of lx1*ly1*lz1,lelv points for an element of > tensor. > And so, the global functions like col3 of glsc can not be applied. > At least I don't understand how. > > To bypass this, I copied comp_sij to my usr file and changed the order of > last indices in sij > real sij(lx1*ly1*lz1,lelv,nij) > > > After that I use the followng code to compute viscous dissipation > > nij = 3 > if (if3d) nij=6 > > call il_comp_sij(sij,nij,vx,vy,vz,ur,us,ut,vr,vs,vt,wr,ws,wt) > > call rzero(il_dissip,lv) > > do j=1,nij > call col3(sij2,sij(1,1,1,1,j),sij(1,1,1,1,j),nxyz) > il_dissip=il_dissip+sij2 > if (if3d .and. j.ge.4) then > il_dissip=il_dissip+sij2 > end if > if ( .not. if3d .and. j.eq.3 ) then > il_dissip=il_dissip+sij2 > end if > il_dissip=2*param(2)*il_dissip > enddo > > Is is a fair way to compute dissipation? > > Thanks, Ilias > > > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Mon Nov 9 08:18:43 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 9 Nov 2015 06:18:43 -0800 Subject: [Nek5000-users] Newest version of svn repo (1088) does not compile Message-ID: Paul, Thanks. I had not updated my makenek from the repo. When I did update my makenek then the newest version of nek5000 did compile. Janet ------------------------------ Message: 2 Date: Fri, 6 Nov 2015 20:28:45 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: Re: [Nek5000-users] Newest version of svn repo (1088) does not compile Message-ID: Content-Type: text/plain; charset="iso-8859-1" Hi Janet, My fault... we're working on it on this end. Did you update your "makenek" from the repo? Paul On Fri, Nov 6, 2015 at 11:45 AM, Janet Scheel wrote: > Just an FYI for those of you who tend to do svn updates fairly often for > nek5000: > > The svn version 1088 does not compile correctly. I found if I reverted to > version 1078 it did compile correctly. > > Janet > > -- > Janet Scheel > Associate Professor of Physics > Occidental College > 1600 Campus Road, M21 > Los Angeles, CA 90041 > (323) 259-2777 > jscheel at oxy.edu > -- Janet Scheel Associate Professor of Physics Occidental College 1600 Campus Road, M21 Los Angeles, CA 90041 (323) 259-2777 jscheel at oxy.edu -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Mon Nov 9 09:34:48 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 9 Nov 2015 16:34:48 +0100 Subject: [Nek5000-users] dissipation In-Reply-To: References: Message-ID: An afterquesion; Am I right that I use the obtained field of dissipation for hpts output through call opcopy(vx,vy,pr,sij2,vy,vz) hpts() ? Thank you, Ilias On Mon, Nov 9, 2015 at 5:03 AM, wrote: > > Hi Ilias, > > > How about: > > subroutine dissipation(sij2) > > > include 'SIZE' > > include 'TOTAL' > > > real sij2(lx1*ly1*lz1,lelt) > > > common /scrns/ sij (lx1*ly1*lz1,6,lelv) > > > parameter (lr=lx1*ly1*lz1) > > common /scruz/ ur(lr),us(lr),ut(lr) > > $ , vr(lr),vs(lr),vt(lr) > > $ , wr(lr),ws(lr),wt(lr) > > > integer e > > > nij = 3+3*(ndim-2) > > > call comp_sij(sij,nij,vx,vy,vz,ur,us,ut,vr,vs,vt,wr,ws,wt) > > > nxyz = nx1*ny1*nz1 > > n = nx1*ny1*nz1*nelv > > > call rzero(sij2,n) > > > do e=1,nelv > > do j=1,nij > > do i=1,nxyz > > sij2(i,e)=sij2(i,e)+sij(i,j,e)**2 > > enddo > > enddo > > enddo > > > return > > end > > ? > > This should function correctly. > > Best, Paul > > > ------------------------------ > *From:* nek5000-users-bounces at lists.mcs.anl.gov [ > nek5000-users-bounces at lists.mcs.anl.gov] on behalf of > nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] > *Sent:* Sunday, November 08, 2015 4:37 PM > *To:* nek5000-users at lists.mcs.anl.gov > *Subject:* [Nek5000-users] dissipation > > Dear neks, > I am trying to user comp_sij to compute dissipation in every point of the > domain. > The data storage is organised according to > real sij(lx1*ly1*lz1,nij,lelv) > so the index of a component of tensor > precedes number of element, and > the is no continuous block of lx1*ly1*lz1,lelv points for an element of > tensor. > And so, the global functions like col3 of glsc can not be applied. > At least I don't understand how. > > To bypass this, I copied comp_sij to my usr file and changed the order of > last indices in sij > real sij(lx1*ly1*lz1,lelv,nij) > > > After that I use the followng code to compute viscous dissipation > > nij = 3 > if (if3d) nij=6 > > call il_comp_sij(sij,nij,vx,vy,vz,ur,us,ut,vr,vs,vt,wr,ws,wt) > > call rzero(il_dissip,lv) > > do j=1,nij > call col3(sij2,sij(1,1,1,1,j),sij(1,1,1,1,j),nxyz) > il_dissip=il_dissip+sij2 > if (if3d .and. j.ge.4) then > il_dissip=il_dissip+sij2 > end if > if ( .not. if3d .and. j.eq.3 ) then > il_dissip=il_dissip+sij2 > end if > il_dissip=2*param(2)*il_dissip > enddo > > Is is a fair way to compute dissipation? > > Thanks, Ilias > > > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Mon Nov 9 09:40:50 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 9 Nov 2015 15:40:50 +0000 Subject: [Nek5000-users] dissipation In-Reply-To: References: , Message-ID: Hi Ilias, opcopy is for vector to vector copies - so that would not be correct usage. I would just look carefully at the hpts constructs if you want to interpolate your dissipation. Paul ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Monday, November 09, 2015 9:34 AM To: nek5000-users at lists.mcs.anl.gov Subject: Re: [Nek5000-users] dissipation An afterquesion; Am I right that I use the obtained field of dissipation for hpts output through call opcopy(vx,vy,pr,sij2,vy,vz) hpts() ? Thank you, Ilias On Mon, Nov 9, 2015 at 5:03 AM, > wrote: Hi Ilias, How about: subroutine dissipation(sij2) include 'SIZE' include 'TOTAL' real sij2(lx1*ly1*lz1,lelt) common /scrns/ sij (lx1*ly1*lz1,6,lelv) parameter (lr=lx1*ly1*lz1) common /scruz/ ur(lr),us(lr),ut(lr) $ , vr(lr),vs(lr),vt(lr) $ , wr(lr),ws(lr),wt(lr) integer e nij = 3+3*(ndim-2) call comp_sij(sij,nij,vx,vy,vz,ur,us,ut,vr,vs,vt,wr,ws,wt) nxyz = nx1*ny1*nz1 n = nx1*ny1*nz1*nelv call rzero(sij2,n) do e=1,nelv do j=1,nij do i=1,nxyz sij2(i,e)=sij2(i,e)+sij(i,j,e)**2 enddo enddo enddo return end ? This should function correctly. Best, Paul ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Sunday, November 08, 2015 4:37 PM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] dissipation Dear neks, I am trying to user comp_sij to compute dissipation in every point of the domain. The data storage is organised according to real sij(lx1*ly1*lz1,nij,lelv) so the index of a component of tensor precedes number of element, and the is no continuous block of lx1*ly1*lz1,lelv points for an element of tensor. And so, the global functions like col3 of glsc can not be applied. At least I don't understand how. To bypass this, I copied comp_sij to my usr file and changed the order of last indices in sij real sij(lx1*ly1*lz1,lelv,nij) After that I use the followng code to compute viscous dissipation nij = 3 if (if3d) nij=6 call il_comp_sij(sij,nij,vx,vy,vz,ur,us,ut,vr,vs,vt,wr,ws,wt) call rzero(il_dissip,lv) do j=1,nij call col3(sij2,sij(1,1,1,1,j),sij(1,1,1,1,j),nxyz) il_dissip=il_dissip+sij2 if (if3d .and. j.ge.4) then il_dissip=il_dissip+sij2 end if if ( .not. if3d .and. j.eq.3 ) then il_dissip=il_dissip+sij2 end if il_dissip=2*param(2)*il_dissip enddo Is is a fair way to compute dissipation? Thanks, Ilias _______________________________________________ Nek5000-users mailing list Nek5000-users at lists.mcs.anl.gov https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Mon Nov 9 10:22:38 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 9 Nov 2015 17:22:38 +0100 Subject: [Nek5000-users] dissipation In-Reply-To: References: Message-ID: Thank you, Paul Do you mean that opcopy does unnecessary work of copying three components of vector, when I need really output of only only scalar? But as a result of call opcopy(vx,vy,vx,sij2,vy,vz) call hpts() It is important for me to understand now if I will anyway have in the first column the sij2? Or you mean that vx and sij2 structures are different, so opcopy is not correct and I will not have sij2 in the first column? Thank you, Ilias I hoped It would be easier... But if I am right call gradm1(dtdx,dtdy,dtdz,t) or call comp_vort3(v1,v2,v3,vx,vy,vz) preserves vector structure, so it can be used with opcopy. So may be there is a way to construct dissipation which can also be used On Mon, Nov 9, 2015 at 4:40 PM, wrote: > > Hi Ilias, > > opcopy is for vector to vector copies - so that would not be correct usage. > > I would just look carefully at the hpts constructs if you want to > interpolate your > dissipation. > > Paul > > ------------------------------ > *From:* nek5000-users-bounces at lists.mcs.anl.gov [ > nek5000-users-bounces at lists.mcs.anl.gov] on behalf of > nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] > *Sent:* Monday, November 09, 2015 9:34 AM > *To:* nek5000-users at lists.mcs.anl.gov > *Subject:* Re: [Nek5000-users] dissipation > > An afterquesion; > Am I right that I use the obtained field of dissipation > for hpts output through > call opcopy(vx,vy,pr,sij2,vy,vz) > hpts() > ? > > Thank you, Ilias > > > On Mon, Nov 9, 2015 at 5:03 AM, wrote: > >> >> Hi Ilias, >> >> >> How about: >> >> subroutine dissipation(sij2) >> >> >> include 'SIZE' >> >> include 'TOTAL' >> >> >> real sij2(lx1*ly1*lz1,lelt) >> >> >> common /scrns/ sij (lx1*ly1*lz1,6,lelv) >> >> >> parameter (lr=lx1*ly1*lz1) >> >> common /scruz/ ur(lr),us(lr),ut(lr) >> >> $ , vr(lr),vs(lr),vt(lr) >> >> $ , wr(lr),ws(lr),wt(lr) >> >> >> integer e >> >> >> nij = 3+3*(ndim-2) >> >> >> call comp_sij(sij,nij,vx,vy,vz,ur,us,ut,vr,vs,vt,wr,ws,wt) >> >> >> nxyz = nx1*ny1*nz1 >> >> n = nx1*ny1*nz1*nelv >> >> >> call rzero(sij2,n) >> >> >> do e=1,nelv >> >> do j=1,nij >> >> do i=1,nxyz >> >> sij2(i,e)=sij2(i,e)+sij(i,j,e)**2 >> >> enddo >> >> enddo >> >> enddo >> >> >> return >> >> end >> >> ? >> >> This should function correctly. >> >> Best, Paul >> >> >> ------------------------------ >> *From:* nek5000-users-bounces at lists.mcs.anl.gov [ >> nek5000-users-bounces at lists.mcs.anl.gov] on behalf of >> nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] >> *Sent:* Sunday, November 08, 2015 4:37 PM >> *To:* nek5000-users at lists.mcs.anl.gov >> *Subject:* [Nek5000-users] dissipation >> >> Dear neks, >> I am trying to user comp_sij to compute dissipation in every point of the >> domain. >> The data storage is organised according to >> real sij(lx1*ly1*lz1,nij,lelv) >> so the index of a component of tensor >> precedes number of element, and >> the is no continuous block of lx1*ly1*lz1,lelv points for an element of >> tensor. >> And so, the global functions like col3 of glsc can not be applied. >> At least I don't understand how. >> >> To bypass this, I copied comp_sij to my usr file and changed the order of >> last indices in sij >> real sij(lx1*ly1*lz1,lelv,nij) >> >> >> After that I use the followng code to compute viscous dissipation >> >> nij = 3 >> if (if3d) nij=6 >> >> call il_comp_sij(sij,nij,vx,vy,vz,ur,us,ut,vr,vs,vt,wr,ws,wt) >> >> call rzero(il_dissip,lv) >> >> do j=1,nij >> call col3(sij2,sij(1,1,1,1,j),sij(1,1,1,1,j),nxyz) >> il_dissip=il_dissip+sij2 >> if (if3d .and. j.ge.4) then >> il_dissip=il_dissip+sij2 >> end if >> if ( .not. if3d .and. j.eq.3 ) then >> il_dissip=il_dissip+sij2 >> end if >> il_dissip=2*param(2)*il_dissip >> enddo >> >> Is is a fair way to compute dissipation? >> >> Thanks, Ilias >> >> >> >> _______________________________________________ >> Nek5000-users mailing list >> Nek5000-users at lists.mcs.anl.gov >> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users >> >> > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Tue Nov 10 03:31:50 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 10 Nov 2015 10:31:50 +0100 Subject: [Nek5000-users] velocity spikes at the corner each process sub-domain Message-ID: Hi neks, I'm using nek5000 with the low-mach approximation and the CVODE implicit ode solver for a simple one step combustion problem in 2D configuration. I have two issue: a) During a parallel run using 24 or less processors on a UNIX system, I have big velocity spikes at the corner each process sub-domain. In serial runs this is not present using the same mpi compilers ( I'm using openmpi-1.5.3). b) I'm not able to insert a zero-thickness insulated wall in the domain, species seem to penetrate and diffuse into it. I use genbox for the mesh and the conditions at the block boundaries I used are the following: W for the velocity and I for tempreature and scalars. Any suggestion will be great! Regards Pasquale -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Wed Nov 11 07:17:14 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 11 Nov 2015 14:17:14 +0100 Subject: [Nek5000-users] what is? Message-ID: Hi neks, during genmap execution i have this output: done: 0.0% done: 1.0% done: 2.0% done: 3.0% done: 4.0% done: 5.0% done: 6.0% done: 7.0% done: 8.0% done: 9.0% done: 10.0% done: 11.0% done: 12.0% done: 13.0% done: 14.0% not connected 1 2 1 0 done: 15.0% done: 16.0% done: 17.0% done: 18.0% done: 19.0% done: 20.0% done: 21.0% done: 22.0% done: 23.0% done: 24.0% done: 25.0% i have copied only till 25%, but sometime i have "not connected x x x x ", can someone tell me what does it mean? I read relative genmap.f file, but i didn't undesrtand. However at least it complete the command and i have the final .map file but i have errors during compilation, but i don't know if they aere related to these ones errors or they are memory issues. From nek5000-users at lists.mcs.anl.gov Wed Nov 11 08:29:18 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 11 Nov 2015 14:29:18 +0000 Subject: [Nek5000-users] what is? In-Reply-To: References: Message-ID: This just gives you an indication of how much of the partitioning is completed (which is done by recursive spectral bisection). Disconnected subgraphs are ok if the numbers in the row are small (<10). If they are larger, it indicates some kind of problem. The case will still run in any case, but performance could be suboptimal for large processor counts if you have large disconnected subgraphs because there would be more communication than expected. genmap takes steps to correct disconnected subgraphs but is not always successful. If it can't correct, it moves on and you still have a load-balanced partition. Paul ________________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Wednesday, November 11, 2015 7:17 AM To: nek5000-users at lists.mcs .anl.gov Subject: [Nek5000-users] what is? Hi neks, during genmap execution i have this output: done: 0.0% done: 1.0% done: 2.0% done: 3.0% done: 4.0% done: 5.0% done: 6.0% done: 7.0% done: 8.0% done: 9.0% done: 10.0% done: 11.0% done: 12.0% done: 13.0% done: 14.0% not connected 1 2 1 0 done: 15.0% done: 16.0% done: 17.0% done: 18.0% done: 19.0% done: 20.0% done: 21.0% done: 22.0% done: 23.0% done: 24.0% done: 25.0% i have copied only till 25%, but sometime i have "not connected x x x x ", can someone tell me what does it mean? I read relative genmap.f file, but i didn't undesrtand. However at least it complete the command and i have the final .map file but i have errors during compilation, but i don't know if they aere related to these ones errors or they are memory issues. _______________________________________________ Nek5000-users mailing list Nek5000-users at lists.mcs.anl.gov https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users From nek5000-users at lists.mcs.anl.gov Wed Nov 11 08:40:55 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 11 Nov 2015 14:40:55 +0000 Subject: [Nek5000-users] FW: Nek and Moab In-Reply-To: <00cd01d11bd3$fcd4a9e0$f67dfda0$@msn.com> References: <00cd01d11bd3$fcd4a9e0$f67dfda0$@msn.com> Message-ID: I'm resending the same email from my ORNL account. Please somebody let me know what the problem is. Thanks, emilian From: Emilian Popov [mailto:emilian_popov at msn.com] Sent: Tuesday, November 10, 2015 11:22 AM To: 'nek5000-users at lists.mcs.anl.gov' Subject: Nek and Moab Dear Neks, I'm trying to run Nek with Moab. The code seems to compile, but when I run it, it aborts with error: this version of Nek not compiled with Moab What have I done wrong? Thx, emilian -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Wed Nov 11 09:25:16 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 11 Nov 2015 09:25:16 -0600 Subject: [Nek5000-users] FW: Nek and Moab In-Reply-To: References: <00cd01d11bd3$fcd4a9e0$f67dfda0$@msn.com> Message-ID: You need to include the following in makenek to get started. PPLIST="MOAB" MOAB_DIR="PATH_TO_MOAB_INSTALLATION" Then if using GNU compilers, add USR_LFLAGS="-lmpi_cxx -lstdc++" or with Intel, USR_LFLAGS="-cxxlib" That should be sufficient to get started I think. In the future, always send detailed error messages if you want us to help you quickly. Vijay On Wed, Nov 11, 2015 at 8:40 AM, wrote: > I?m resending the same email from my ORNL account. Please somebody let me > know what the problem is. > > > > Thanks, > > emilian > > > > From: Emilian Popov [mailto:emilian_popov at msn.com] > Sent: Tuesday, November 10, 2015 11:22 AM > To: 'nek5000-users at lists.mcs.anl.gov' > Subject: Nek and Moab > > > > Dear Neks, > > > > I?m trying to run Nek with Moab. The code seems to compile, but when I run > it, it aborts with error: this version of Nek not compiled with Moab > > > > What have I done wrong? > > > > Thx, emilian > > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > From nek5000-users at lists.mcs.anl.gov Thu Nov 12 04:17:34 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 12 Nov 2015 11:17:34 +0100 Subject: [Nek5000-users] Format bug in navier4.f: format/variable-type mismatch Message-ID: Hi all, I recently stumbled upon the following error message with Nek: forrtl: severe (61): format/variable-type mismatch, unit 6, file /proc/22947/fd/1 This seems to be due to a format mixup in navier4.f. Switching the position of '2i4' and 'a6' in Line 1092: 1 format(i9,'proj_ortho: ',2i4,a6,' Detect rank deficiency:' $ ,1p2e12.4) solved the problem for me. Cheers, Jan From nek5000-users at lists.mcs.anl.gov Fri Nov 13 09:13:54 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 13 Nov 2015 16:13:54 +0100 Subject: [Nek5000-users] Trouble with stress formulation Message-ID: Hi all I am having some trouble when activating the IFSTRS flag. The eigenvalues seem to blow up, and the mesh changes. The warning message is just the beginning of a long list of mismatches I suspect is caused by the unwanted change in the mesh topology. Below is a part of the output with and without activating the stress-fromulation. ******** IFSTRS = TRUE ******************* Estimated eigenvalues EIGAA = 1.83548835643387 EIGGA = NaN EIGAE = 0.314292232447382 EIGAS = 1.426923754586004E-002 EIGGE = NaN EIGGS = 2.00000000000000 verify mesh topology -0.643805403845627 4.96000004000001 Xrange -1.75000000000000 1.75000000000000 Yrange 0.000000000000000E+000 1.50000000000000 Zrange WARNING1 Element mesh mismatch at: i,j,k,ie: 8 1 2 3836 ******** IFSTRS = FALSE ******************* Estimated eigenvalues EIGAA = 1.86445005547390 EIGGA = 161789138.855617 EIGAE = 0.401177329567469 EIGAS = 1.793099473620650E-002 EIGGE = 161789138.855617 EIGGS = 2.00000000000000 verify mesh topology -1.276863900000000E-008 4.96000004000001 Xrange -1.75000000000000 1.75000000000000 Yrange 0.000000000000000E+000 1.50000000000000 Zrange done :: verify mesh topology ************************************************* Any help will be appreciated. Regards Magnus -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Tue Nov 10 10:22:23 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 10 Nov 2015 11:22:23 -0500 Subject: [Nek5000-users] Nek and Moab Message-ID: Dear Neks, I'm trying to run Nek with Moab. The code seems to compile, but when I run it, it aborts with error: this version of Nek not compiled with Moab What have a done wrong? Thx, emilian -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Fri Nov 13 12:37:45 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 13 Nov 2015 18:37:45 +0000 Subject: [Nek5000-users] Nek and Moab In-Reply-To: References: Message-ID: Dear Emilian, I might be able to help sort this out. Can you contact me off-list? Best, Paul ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Tuesday, November 10, 2015 10:22 AM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Nek and Moab Dear Neks, I?m trying to run Nek with Moab. The code seems to compile, but when I run it, it aborts with error: this version of Nek not compiled with Moab What have a done wrong? Thx, emilian -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Mon Nov 16 13:19:23 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 16 Nov 2015 19:19:23 +0000 Subject: [Nek5000-users] Full restart with a different time step Message-ID: Dear Neks, I use full_restart option for my current simulations. One simulation crashed after restarting for some time with the Courant number going crazy. I suppose I will need to reduce the time step size. Basically, I want to continue my simulation using full_restart option but with a smaller time step size. Somehow I remember I saw one post saying one shouldn't change the time step size when using the full_restart option. I'm just thinking of a way to work around the problem. So does anyone have ideas on how I can interpolate the flow fields between two time steps or any other ways to fully restart a simulation with a different time step size? Thank you very much in advance for any suggestions or comments. Best regards, Tony -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Tue Nov 17 02:27:47 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 17 Nov 2015 09:27:47 +0100 Subject: [Nek5000-users] Full restart with a different time step In-Reply-To: References: Message-ID: Hi For full_restart you should have four velocity fields and you could use nek5000 as a preprocessing tool. You can simply load all 4 fields in userchk, perform some kind of interpolation in time and finally write down new restart files. The question is how to deal with pressure. One solution would be simply interpolate pressure saved in the files. The other would be to run modified version of pressure solver with uzawa turned on to get pressure corresponding to interpolated velocity field. We've played with uzawa for pn-pn-2 and we use our version of full_restart, but we haven't done this kind of modified restart, so I don't know what is better. On the other hand you could restart nek with your current checkpoints and dt corresponding to the old simulation, and later (after 4th step) gradually decrease time step. I guess this would be the simplest and probably best solution. If time resolution in old simulation was insufficient, your restart files do not give correct evolution of the field and interpolation in time would not remove it. Anyhow you introduce some initial error to simulation. In this case I probably would simply play with dt keeping it constant during first 4 steps and gradually decreasing later. Regards Adam On 2015-11-16 20:19, nek5000-users at lists.mcs.anl.gov wrote: > > Dear Neks, > > > I use full_restart option for my current simulations. One simulation > crashed after restarting for some time with the Courant number going > crazy. I suppose I will need to reduce the time step size. > > > Basically, I want to continue my simulation using full_restart option > but with a smaller time step size. Somehow I remember I saw one post > saying one shouldn't change the time step size when using the > full_restart option. I'm just thinking of a way to work around the > problem. > > > So does anyone have ideas on how I can interpolate the flow fields > between two time steps or any other ways to fully restart a simulation > with a different time step size? > > > Thank you very much in advance for any suggestions or comments. > > > Best regards, > > Tony > > > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Tue Nov 17 15:20:58 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 17 Nov 2015 21:20:58 +0000 Subject: [Nek5000-users] New Webpage and Documentation Message-ID: Hi nek5000 users, As you may have noticed we have a new webpage and a documentation update where we gathered the information on the previous pages. As last week the documentation is accessible in both pdf and browsable form. Some info may have been missed either by accident or since it seemed redundant. Therefore we welcome your requests and observations in order to improve the current status. The old page is now hosted statically on the new page but as Aleks pointed out the search button doesn't work anymore. The reason for that is that our systems people couldn't mirror the old page and could barely give us the old page statically. I realize this may be highly inconvenient to some but the best solution we can see such that we do not short circuit the progress: 1. If there is anything you are looking for and is not available please let us know and we shall update it 2. We try to adapt the search button on the new page to allow users to navigate as they were used to, also the browsable edition of the documentation should be of help 3. The old page source together with the database can be provided to whoever sees any use to it. Also you may have noticed we link to most of the centers that have a high number of users, and also promote work performed with Nek5000. If you would like to be added either by lining to your homepage or just by citing your work in the Gallery please let us know. Best, Oana From nek5000-users at lists.mcs.anl.gov Tue Nov 17 17:24:07 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 17 Nov 2015 23:24:07 +0000 Subject: [Nek5000-users] Full restart with a different time step In-Reply-To: References: Message-ID: Hi Adam, Thank you very much for your comments. I actually thought about loading the 4 restart files in nek and do interpolation between two time steps. But I agree with you that I would introduce some error to the flow fields for restarting. In my case, I want to keep the data of the old simulation and continue from the current checkpoints, so I don't want to ruin the accuracy of the flow fields. Like you said, changing \dt after restarting from the current flow fields maight be the simplest solution. However, I have one concern about this. When we use 3rd order accuracy in time, the solver would use the history information from the last few time steps and that's why we need to restart the simulation with 4 continuous flow fields. If I change the time step size in the middle of the simulation, how much would it affect the solving process? What do you think? Best regards, Tony ________________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov on behalf of nek5000-users-request at lists.mcs.anl.gov Sent: 17 November 2015 18:00 To: nek5000-users at lists.mcs.anl.gov Subject: Nek5000-users Digest, Vol 81, Issue 15 Send Nek5000-users mailing list submissions to nek5000-users at lists.mcs.anl.gov To subscribe or unsubscribe via the World Wide Web, visit https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users or, via email, send a message with subject or body 'help' to nek5000-users-request at lists.mcs.anl.gov You can reach the person managing the list at nek5000-users-owner at lists.mcs.anl.gov When replying, please edit your Subject line so it is more specific than "Re: Contents of Nek5000-users digest..." Today's Topics: 1. Full restart with a different time step (nek5000-users at lists.mcs.anl.gov) 2. Re: Full restart with a different time step (nek5000-users at lists.mcs.anl.gov) ---------------------------------------------------------------------- Message: 1 Date: Mon, 16 Nov 2015 19:19:23 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: [Nek5000-users] Full restart with a different time step Message-ID: Content-Type: text/plain; charset="iso-8859-1" Dear Neks, I use full_restart option for my current simulations. One simulation crashed after restarting for some time with the Courant number going crazy. I suppose I will need to reduce the time step size. Basically, I want to continue my simulation using full_restart option but with a smaller time step size. Somehow I remember I saw one post saying one shouldn't change the time step size when using the full_restart option. I'm just thinking of a way to work around the problem. So does anyone have ideas on how I can interpolate the flow fields between two time steps or any other ways to fully restart a simulation with a different time step size? Thank you very much in advance for any suggestions or comments. Best regards, Tony -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Message: 2 Date: Tue, 17 Nov 2015 09:27:47 +0100 From: nek5000-users at lists.mcs.anl.gov To: nek5000-users at lists.mcs.anl.gov Subject: Re: [Nek5000-users] Full restart with a different time step Message-ID: Content-Type: text/plain; charset="windows-1252"; Format="flowed" Hi For full_restart you should have four velocity fields and you could use nek5000 as a preprocessing tool. You can simply load all 4 fields in userchk, perform some kind of interpolation in time and finally write down new restart files. The question is how to deal with pressure. One solution would be simply interpolate pressure saved in the files. The other would be to run modified version of pressure solver with uzawa turned on to get pressure corresponding to interpolated velocity field. We've played with uzawa for pn-pn-2 and we use our version of full_restart, but we haven't done this kind of modified restart, so I don't know what is better. On the other hand you could restart nek with your current checkpoints and dt corresponding to the old simulation, and later (after 4th step) gradually decrease time step. I guess this would be the simplest and probably best solution. If time resolution in old simulation was insufficient, your restart files do not give correct evolution of the field and interpolation in time would not remove it. Anyhow you introduce some initial error to simulation. In this case I probably would simply play with dt keeping it constant during first 4 steps and gradually decreasing later. Regards Adam On 2015-11-16 20:19, nek5000-users at lists.mcs.anl.gov wrote: > > Dear Neks, > > > I use full_restart option for my current simulations. One simulation > crashed after restarting for some time with the Courant number going > crazy. I suppose I will need to reduce the time step size. > > > Basically, I want to continue my simulation using full_restart option > but with a smaller time step size. Somehow I remember I saw one post > saying one shouldn't change the time step size when using the > full_restart option. I'm just thinking of a way to work around the > problem. > > > So does anyone have ideas on how I can interpolate the flow fields > between two time steps or any other ways to fully restart a simulation > with a different time step size? > > > Thank you very much in advance for any suggestions or comments. > > > Best regards, > > Tony > > > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ _______________________________________________ Nek5000-users mailing list Nek5000-users at lists.mcs.anl.gov https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users End of Nek5000-users Digest, Vol 81, Issue 15 ********************************************* From nek5000-users at lists.mcs.anl.gov Tue Nov 17 17:29:40 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 18 Nov 2015 00:29:40 +0100 Subject: [Nek5000-users] Full restart with a different time step In-Reply-To: References: Message-ID: Hi, when you change the time step during a simulation, then there is no problem; the EXT/BDF coefficients are adjusted accordingly, so you will keep the accuracy/order even with varying time step. The reason why you cannot do that during restart (i.e. during the first 4 steps) is because the stored fields are at a given time (at which they were stored, yielding a certain dt), so you cannot assign them a different time or dt. The solution in memory is then overwritten with the data loaded from disk, until all lag arrays are correctly filled. But as soon as the simulation is not reading from the disk any longer, then you can go ahead and change the time step. Best, PHilipp On 2015-11-18 00:24, nek5000-users at lists.mcs.anl.gov wrote: > Hi Adam, > > Thank you very much for your comments. > > I actually thought about loading the 4 restart files in nek and do > interpolation between two time steps. But I agree with you that I > would introduce some error to the flow fields for restarting. In my > case, I want to keep the data of the old simulation and continue from > the current checkpoints, so I don't want to ruin the accuracy of the > flow fields. > > Like you said, changing \dt after restarting from the current flow > fields maight be the simplest solution. However, I have one concern > about this. When we use 3rd order accuracy in time, the solver would > use the history information from the last few time steps and that's > why we need to restart the simulation with 4 continuous flow fields. > If I change the time step size in the middle of the simulation, how > much would it affect the solving process? What do you think? > > Best regards, Tony > > ________________________________________ From: > nek5000-users-bounces at lists.mcs.anl.gov > on behalf of > nek5000-users-request at lists.mcs.anl.gov > Sent: 17 November 2015 > 18:00 To: nek5000-users at lists.mcs.anl.gov Subject: Nek5000-users > Digest, Vol 81, Issue 15 > > Send Nek5000-users mailing list submissions to > nek5000-users at lists.mcs.anl.gov > > To subscribe or unsubscribe via the World Wide Web, visit > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users or, via > email, send a message with subject or body 'help' to > nek5000-users-request at lists.mcs.anl.gov > > You can reach the person managing the list at > nek5000-users-owner at lists.mcs.anl.gov > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Nek5000-users digest..." > > > Today's Topics: > > 1. Full restart with a different time step > (nek5000-users at lists.mcs.anl.gov) 2. Re: Full restart with a > different time step (nek5000-users at lists.mcs.anl.gov) > > > ---------------------------------------------------------------------- > > Message: 1 Date: Mon, 16 Nov 2015 19:19:23 +0000 From: > nek5000-users at lists.mcs.anl.gov To: > "nek5000-users at lists.mcs.anl.gov" > Subject: [Nek5000-users] Full restart with a different time step > Message-ID: > > Content-Type: text/plain; charset="iso-8859-1" > > Dear Neks, > > > I use full_restart option for my current simulations. One simulation > crashed after restarting for some time with the Courant number going > crazy. I suppose I will need to reduce the time step size. > > > Basically, I want to continue my simulation using full_restart option > but with a smaller time step size. Somehow I remember I saw one post > saying one shouldn't change the time step size when using the > full_restart option. I'm just thinking of a way to work around the > problem. > > > So does anyone have ideas on how I can interpolate the flow fields > between two time steps or any other ways to fully restart a > simulation with a different time step size? > > > Thank you very much in advance for any suggestions or comments. > > > Best regards, > > Tony -------------- next part -------------- An HTML attachment was > scrubbed... URL: > > > ------------------------------ > > Message: 2 Date: Tue, 17 Nov 2015 09:27:47 +0100 From: > nek5000-users at lists.mcs.anl.gov To: nek5000-users at lists.mcs.anl.gov > Subject: Re: [Nek5000-users] Full restart with a different time step > Message-ID: > > Content-Type: text/plain; charset="windows-1252"; Format="flowed" > > Hi > > For full_restart you should have four velocity fields and you could > use nek5000 as a preprocessing tool. You can simply load all 4 fields > in userchk, perform some kind of interpolation in time and finally > write down new restart files. The question is how to deal with > pressure. One solution would be simply interpolate pressure saved in > the files. The other would be to run modified version of pressure > solver with uzawa turned on to get pressure corresponding to > interpolated velocity field. We've played with uzawa for pn-pn-2 and > we use our version of full_restart, but we haven't done this kind of > modified restart, so I don't know what is better. On the other hand > you could restart nek with your current checkpoints and dt > corresponding to the old simulation, and later (after 4th step) > gradually decrease time step. I guess this would be the simplest and > probably best solution. If time resolution in old simulation was > insufficient, your restart files do not give correct evolution of the > field and interpolation in time would not remove it. Anyhow you > introduce some initial error to simulation. In this case I probably > would simply play with dt keeping it constant during first 4 steps > and gradually decreasing later. Regards Adam > > On 2015-11-16 20:19, nek5000-users at lists.mcs.anl.gov wrote: >> >> Dear Neks, >> >> >> I use full_restart option for my current simulations. One >> simulation crashed after restarting for some time with the Courant >> number going crazy. I suppose I will need to reduce the time step >> size. >> >> >> Basically, I want to continue my simulation using full_restart >> option but with a smaller time step size. Somehow I remember I saw >> one post saying one shouldn't change the time step size when using >> the full_restart option. I'm just thinking of a way to work around >> the problem. >> >> >> So does anyone have ideas on how I can interpolate the flow fields >> between two time steps or any other ways to fully restart a >> simulation with a different time step size? >> >> >> Thank you very much in advance for any suggestions or comments. >> >> >> Best regards, >> >> Tony >> >> >> >> _______________________________________________ Nek5000-users >> mailing list Nek5000-users at lists.mcs.anl.gov >> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > -------------- next part -------------- An HTML attachment was > scrubbed... URL: > > > ------------------------------ > > _______________________________________________ Nek5000-users mailing > list Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > > End of Nek5000-users Digest, Vol 81, Issue 15 > ********************************************* > _______________________________________________ Nek5000-users mailing > list Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > From nek5000-users at lists.mcs.anl.gov Wed Nov 18 15:05:10 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 18 Nov 2015 21:05:10 +0000 Subject: [Nek5000-users] Full restart with a different time step In-Reply-To: References: Message-ID: Hi Philipp, That makes sense. Thank you very much!! Best regards, Tony ________________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov on behalf of nek5000-users-request at lists.mcs.anl.gov Sent: 18 November 2015 18:00 To: nek5000-users at lists.mcs.anl.gov Subject: Nek5000-users Digest, Vol 81, Issue 16 Send Nek5000-users mailing list submissions to nek5000-users at lists.mcs.anl.gov To subscribe or unsubscribe via the World Wide Web, visit https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users or, via email, send a message with subject or body 'help' to nek5000-users-request at lists.mcs.anl.gov You can reach the person managing the list at nek5000-users-owner at lists.mcs.anl.gov When replying, please edit your Subject line so it is more specific than "Re: Contents of Nek5000-users digest..." Today's Topics: 1. New Webpage and Documentation (nek5000-users at lists.mcs.anl.gov) 2. Re: Full restart with a different time step (nek5000-users at lists.mcs.anl.gov) 3. Re: Full restart with a different time step (nek5000-users at lists.mcs.anl.gov) ---------------------------------------------------------------------- Message: 1 Date: Tue, 17 Nov 2015 21:20:58 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: [Nek5000-users] New Webpage and Documentation Message-ID: Content-Type: text/plain; charset="us-ascii" Hi nek5000 users, As you may have noticed we have a new webpage and a documentation update where we gathered the information on the previous pages. As last week the documentation is accessible in both pdf and browsable form. Some info may have been missed either by accident or since it seemed redundant. Therefore we welcome your requests and observations in order to improve the current status. The old page is now hosted statically on the new page but as Aleks pointed out the search button doesn't work anymore. The reason for that is that our systems people couldn't mirror the old page and could barely give us the old page statically. I realize this may be highly inconvenient to some but the best solution we can see such that we do not short circuit the progress: 1. If there is anything you are looking for and is not available please let us know and we shall update it 2. We try to adapt the search button on the new page to allow users to navigate as they were used to, also the browsable edition of the documentation should be of help 3. The old page source together with the database can be provided to whoever sees any use to it. Also you may have noticed we link to most of the centers that have a high number of users, and also promote work performed with Nek5000. If you would like to be added either by lining to your homepage or just by citing your work in the Gallery please let us know. Best, Oana ------------------------------ Message: 2 Date: Tue, 17 Nov 2015 23:24:07 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: Re: [Nek5000-users] Full restart with a different time step Message-ID: Content-Type: text/plain; charset="iso-8859-1" Hi Adam, Thank you very much for your comments. I actually thought about loading the 4 restart files in nek and do interpolation between two time steps. But I agree with you that I would introduce some error to the flow fields for restarting. In my case, I want to keep the data of the old simulation and continue from the current checkpoints, so I don't want to ruin the accuracy of the flow fields. Like you said, changing \dt after restarting from the current flow fields maight be the simplest solution. However, I have one concern about this. When we use 3rd order accuracy in time, the solver would use the history information from the last few time steps and that's why we need to restart the simulation with 4 continuous flow fields. If I change the time step size in the middle of the simulation, how much would it affect the solving process? What do you think? Best regards, Tony ________________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov on behalf of nek5000-users-request at lists.mcs.anl.gov Sent: 17 November 2015 18:00 To: nek5000-users at lists.mcs.anl.gov Subject: Nek5000-users Digest, Vol 81, Issue 15 Send Nek5000-users mailing list submissions to nek5000-users at lists.mcs.anl.gov To subscribe or unsubscribe via the World Wide Web, visit https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users or, via email, send a message with subject or body 'help' to nek5000-users-request at lists.mcs.anl.gov You can reach the person managing the list at nek5000-users-owner at lists.mcs.anl.gov When replying, please edit your Subject line so it is more specific than "Re: Contents of Nek5000-users digest..." Today's Topics: 1. Full restart with a different time step (nek5000-users at lists.mcs.anl.gov) 2. Re: Full restart with a different time step (nek5000-users at lists.mcs.anl.gov) ---------------------------------------------------------------------- Message: 1 Date: Mon, 16 Nov 2015 19:19:23 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: [Nek5000-users] Full restart with a different time step Message-ID: Content-Type: text/plain; charset="iso-8859-1" Dear Neks, I use full_restart option for my current simulations. One simulation crashed after restarting for some time with the Courant number going crazy. I suppose I will need to reduce the time step size. Basically, I want to continue my simulation using full_restart option but with a smaller time step size. Somehow I remember I saw one post saying one shouldn't change the time step size when using the full_restart option. I'm just thinking of a way to work around the problem. So does anyone have ideas on how I can interpolate the flow fields between two time steps or any other ways to fully restart a simulation with a different time step size? Thank you very much in advance for any suggestions or comments. Best regards, Tony -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Message: 2 Date: Tue, 17 Nov 2015 09:27:47 +0100 From: nek5000-users at lists.mcs.anl.gov To: nek5000-users at lists.mcs.anl.gov Subject: Re: [Nek5000-users] Full restart with a different time step Message-ID: Content-Type: text/plain; charset="windows-1252"; Format="flowed" Hi For full_restart you should have four velocity fields and you could use nek5000 as a preprocessing tool. You can simply load all 4 fields in userchk, perform some kind of interpolation in time and finally write down new restart files. The question is how to deal with pressure. One solution would be simply interpolate pressure saved in the files. The other would be to run modified version of pressure solver with uzawa turned on to get pressure corresponding to interpolated velocity field. We've played with uzawa for pn-pn-2 and we use our version of full_restart, but we haven't done this kind of modified restart, so I don't know what is better. On the other hand you could restart nek with your current checkpoints and dt corresponding to the old simulation, and later (after 4th step) gradually decrease time step. I guess this would be the simplest and probably best solution. If time resolution in old simulation was insufficient, your restart files do not give correct evolution of the field and interpolation in time would not remove it. Anyhow you introduce some initial error to simulation. In this case I probably would simply play with dt keeping it constant during first 4 steps and gradually decreasing later. Regards Adam On 2015-11-16 20:19, nek5000-users at lists.mcs.anl.gov wrote: > > Dear Neks, > > > I use full_restart option for my current simulations. One simulation > crashed after restarting for some time with the Courant number going > crazy. I suppose I will need to reduce the time step size. > > > Basically, I want to continue my simulation using full_restart option > but with a smaller time step size. Somehow I remember I saw one post > saying one shouldn't change the time step size when using the > full_restart option. I'm just thinking of a way to work around the > problem. > > > So does anyone have ideas on how I can interpolate the flow fields > between two time steps or any other ways to fully restart a simulation > with a different time step size? > > > Thank you very much in advance for any suggestions or comments. > > > Best regards, > > Tony > > > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ _______________________________________________ Nek5000-users mailing list Nek5000-users at lists.mcs.anl.gov https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users End of Nek5000-users Digest, Vol 81, Issue 15 ********************************************* ------------------------------ Message: 3 Date: Wed, 18 Nov 2015 00:29:40 +0100 From: nek5000-users at lists.mcs.anl.gov To: nek5000-users at lists.mcs.anl.gov Subject: Re: [Nek5000-users] Full restart with a different time step Message-ID: Content-Type: text/plain; charset=windows-1252; format=flowed Hi, when you change the time step during a simulation, then there is no problem; the EXT/BDF coefficients are adjusted accordingly, so you will keep the accuracy/order even with varying time step. The reason why you cannot do that during restart (i.e. during the first 4 steps) is because the stored fields are at a given time (at which they were stored, yielding a certain dt), so you cannot assign them a different time or dt. The solution in memory is then overwritten with the data loaded from disk, until all lag arrays are correctly filled. But as soon as the simulation is not reading from the disk any longer, then you can go ahead and change the time step. Best, PHilipp On 2015-11-18 00:24, nek5000-users at lists.mcs.anl.gov wrote: > Hi Adam, > > Thank you very much for your comments. > > I actually thought about loading the 4 restart files in nek and do > interpolation between two time steps. But I agree with you that I > would introduce some error to the flow fields for restarting. In my > case, I want to keep the data of the old simulation and continue from > the current checkpoints, so I don't want to ruin the accuracy of the > flow fields. > > Like you said, changing \dt after restarting from the current flow > fields maight be the simplest solution. However, I have one concern > about this. When we use 3rd order accuracy in time, the solver would > use the history information from the last few time steps and that's > why we need to restart the simulation with 4 continuous flow fields. > If I change the time step size in the middle of the simulation, how > much would it affect the solving process? What do you think? > > Best regards, Tony > > ________________________________________ From: > nek5000-users-bounces at lists.mcs.anl.gov > on behalf of > nek5000-users-request at lists.mcs.anl.gov > Sent: 17 November 2015 > 18:00 To: nek5000-users at lists.mcs.anl.gov Subject: Nek5000-users > Digest, Vol 81, Issue 15 > > Send Nek5000-users mailing list submissions to > nek5000-users at lists.mcs.anl.gov > > To subscribe or unsubscribe via the World Wide Web, visit > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users or, via > email, send a message with subject or body 'help' to > nek5000-users-request at lists.mcs.anl.gov > > You can reach the person managing the list at > nek5000-users-owner at lists.mcs.anl.gov > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Nek5000-users digest..." > > > Today's Topics: > > 1. Full restart with a different time step > (nek5000-users at lists.mcs.anl.gov) 2. Re: Full restart with a > different time step (nek5000-users at lists.mcs.anl.gov) > > > ---------------------------------------------------------------------- > > Message: 1 Date: Mon, 16 Nov 2015 19:19:23 +0000 From: > nek5000-users at lists.mcs.anl.gov To: > "nek5000-users at lists.mcs.anl.gov" > Subject: [Nek5000-users] Full restart with a different time step > Message-ID: > > Content-Type: text/plain; charset="iso-8859-1" > > Dear Neks, > > > I use full_restart option for my current simulations. One simulation > crashed after restarting for some time with the Courant number going > crazy. I suppose I will need to reduce the time step size. > > > Basically, I want to continue my simulation using full_restart option > but with a smaller time step size. Somehow I remember I saw one post > saying one shouldn't change the time step size when using the > full_restart option. I'm just thinking of a way to work around the > problem. > > > So does anyone have ideas on how I can interpolate the flow fields > between two time steps or any other ways to fully restart a > simulation with a different time step size? > > > Thank you very much in advance for any suggestions or comments. > > > Best regards, > > Tony -------------- next part -------------- An HTML attachment was > scrubbed... URL: > > > ------------------------------ > > Message: 2 Date: Tue, 17 Nov 2015 09:27:47 +0100 From: > nek5000-users at lists.mcs.anl.gov To: nek5000-users at lists.mcs.anl.gov > Subject: Re: [Nek5000-users] Full restart with a different time step > Message-ID: > > Content-Type: text/plain; charset="windows-1252"; Format="flowed" > > Hi > > For full_restart you should have four velocity fields and you could > use nek5000 as a preprocessing tool. You can simply load all 4 fields > in userchk, perform some kind of interpolation in time and finally > write down new restart files. The question is how to deal with > pressure. One solution would be simply interpolate pressure saved in > the files. The other would be to run modified version of pressure > solver with uzawa turned on to get pressure corresponding to > interpolated velocity field. We've played with uzawa for pn-pn-2 and > we use our version of full_restart, but we haven't done this kind of > modified restart, so I don't know what is better. On the other hand > you could restart nek with your current checkpoints and dt > corresponding to the old simulation, and later (after 4th step) > gradually decrease time step. I guess this would be the simplest and > probably best solution. If time resolution in old simulation was > insufficient, your restart files do not give correct evolution of the > field and interpolation in time would not remove it. Anyhow you > introduce some initial error to simulation. In this case I probably > would simply play with dt keeping it constant during first 4 steps > and gradually decreasing later. Regards Adam > > On 2015-11-16 20:19, nek5000-users at lists.mcs.anl.gov wrote: >> >> Dear Neks, >> >> >> I use full_restart option for my current simulations. One >> simulation crashed after restarting for some time with the Courant >> number going crazy. I suppose I will need to reduce the time step >> size. >> >> >> Basically, I want to continue my simulation using full_restart >> option but with a smaller time step size. Somehow I remember I saw >> one post saying one shouldn't change the time step size when using >> the full_restart option. I'm just thinking of a way to work around >> the problem. >> >> >> So does anyone have ideas on how I can interpolate the flow fields >> between two time steps or any other ways to fully restart a >> simulation with a different time step size? >> >> >> Thank you very much in advance for any suggestions or comments. >> >> >> Best regards, >> >> Tony >> >> >> >> _______________________________________________ Nek5000-users >> mailing list Nek5000-users at lists.mcs.anl.gov >> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > -------------- next part -------------- An HTML attachment was > scrubbed... URL: > > > ------------------------------ > > _______________________________________________ Nek5000-users mailing > list Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > > End of Nek5000-users Digest, Vol 81, Issue 15 > ********************************************* > _______________________________________________ Nek5000-users mailing > list Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > ------------------------------ _______________________________________________ Nek5000-users mailing list Nek5000-users at lists.mcs.anl.gov https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users End of Nek5000-users Digest, Vol 81, Issue 16 ********************************************* From nek5000-users at lists.mcs.anl.gov Thu Nov 19 15:33:37 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 19 Nov 2015 15:33:37 -0600 Subject: [Nek5000-users] NEK-MOAB coupling Message-ID: Dear Neks, I think recent version of Nek has some problems to couple Nek with MOAB, I coupled Nek and MOAB before. But with current version of Nek (1090) , it can't compile with MOAB. I attached the makenek output. ( P.S. MOAB version 4.6.3 and compiled with mpi and parallel hdf5 and passed all the tests) Thanks, Fatih -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- makenek - automatic build tool for Nek5000 CMT root /scratch/user/fatihsinan/NEK/trunk/nek/cmt /usr/bin/ld: warning: libquadmath.so.0, needed by /scratch/user/fatihsinan/MOAB/mpi2/mpich-3.2/gcc/lib/libmpi.so, not found (try using -rpath or -rpath-link) ./a.out: error while loading shared libraries: libquadmath.so.0: cannot open shared object file: No such file or directory /usr/bin/ld: warning: libquadmath.so.0, needed by /scratch/user/fatihsinan/MOAB/mpi2/mpich-3.2/gcc/lib/libmpi.so, not found (try using -rpath or -rpath-link) ./a.out: error while loading shared libraries: libquadmath.so.0: cannot open shared object file: No such file or directory generating makefile ... mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/drive.f -o obj/drive.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/drive1.f -o obj/drive1.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/drive2.f -o obj/drive2.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/plan4.f -o obj/plan4.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/bdry.f -o obj/bdry.o /scratch/user/fatihsinan/NEK/trunk/nek/plan4.f:409.25: call add3s2(vext(1,2),vy,vylag,ab0,ab1,ntot) 1 Warning: Array reference at (1) is out of bounds (2 > 1) in dimension 2 /scratch/user/fatihsinan/NEK/trunk/nek/plan4.f:410.34: if(if3d) call add3s2(vext(1,3),vz,vzlag,ab0,ab1,ntot) 1 Warning: Array reference at (1) is out of bounds (3 > 1) in dimension 2 /scratch/user/fatihsinan/NEK/trunk/nek/plan4.f:414.27: call add2s2(vext(1,2),vylag(1,1,1,1,2),ab2,ntot) 1 Warning: Array reference at (1) is out of bounds (2 > 1) in dimension 2 /scratch/user/fatihsinan/NEK/trunk/nek/plan4.f:415.36: if(if3d) call add2s2(vext(1,3),vzlag(1,1,1,1,2),ab2,ntot) 1 Warning: Array reference at (1) is out of bounds (3 > 1) in dimension 2 mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/coef.f -o obj/coef.o /scratch/user/fatihsinan/NEK/trunk/nek/bdry.f:1231.37: cv = vtrans(ix,iy,iz,e,icv)/rho 1 Warning: Array reference at (1) is out of bounds (3 > 2) in dimension 5 /scratch/user/fatihsinan/NEK/trunk/nek/bdry.f:1239.38: lambda = vdiff(ix,iy,iz,e,ilam) 1 Warning: Array reference at (1) is out of bounds (3 > 2) in dimension 5 /scratch/user/fatihsinan/NEK/trunk/nek/bdry.f:1465.20: COMMON /CTMP1/ A1X(LX1),A1Y(LX1),A2X(LX1),A2Y(LX1) 1 Warning: Named COMMON block 'ctmp1' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/bdry.f:1646.20: COMMON /CTMP1/ DRM1(LX1,LX1),DRTM1(LX1,LY1) 1 Warning: Named COMMON block 'ctmp1' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/conduct.f -o obj/conduct.o /scratch/user/fatihsinan/NEK/trunk/nek/bdry.f:1897.20: COMMON /SCRMG/ V1(LX1,LY1,LZ1,LELV) 1 Warning: Named COMMON block 'scrmg' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/drive2.f:1836.20: COMMON /SCRNS/ rw1 (LX1,LY1,LZ1,LELV) 1 Warning: Named COMMON block 'scrns' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/conduct.f:267.20: common /scrns/ ta(lt),tb(lt),h2(lt) 1 Warning: Named COMMON block 'scrns' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/conduct.f:454.20: common /scruz/ ta(lx1,ly1,lz1,lelt) 1 Warning: Named COMMON block 'scruz' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/connect1.f -o obj/connect1.o /scratch/user/fatihsinan/NEK/trunk/nek/connect1.f:719.20: COMMON /CTMP0/ XCB(2,2,2),YCB(2,2,2),ZCB(2,2,2),H(3,3,2),INDX(8) 1 Warning: Named COMMON block 'ctmp0' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/connect2.f -o obj/connect2.o /scratch/user/fatihsinan/NEK/trunk/nek/connect2.f:1653.20: common /scrns/ tc(lx1,ly1,lz1,lelt),td(lx1,ly1,lz1,lelt) 1 Warning: Named COMMON block 'scrns' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/connect2.f:1883.20: COMMON /CTMP0/ VO(LELT),XYZI(3,LELT),CG(3,LELT) 1 Warning: Named COMMON block 'ctmp0' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/dssum.f -o obj/dssum.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/edgec.f -o obj/edgec.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/eigsolv.f -o obj/eigsolv.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/gauss.f -o obj/gauss.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/genxyz.f -o obj/genxyz.o /scratch/user/fatihsinan/NEK/trunk/nek/genxyz.f:329.21: COMMON /SRFCEL/ SUCCES 1 Warning: Named COMMON block 'srfcel' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/navier1.f -o obj/navier1.o /scratch/user/fatihsinan/NEK/trunk/nek/genxyz.f:1298.20: common /ctmp0/ xcb(2,2,2),ycb(2,2,2),zcb(2,2,2),w(ldw) 1 Warning: Named COMMON block 'ctmp0' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/genxyz.f:1349.20: common /ctmp0/ w(ldw,2),zg(3) 1 Warning: Named COMMON block 'ctmp0' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/genxyz.f:1359.20: common /cxyzl/ zgml(lx1,3),jx (lx1*3),jy (lx1*3),jz (lx1*3) 1 Warning: Named COMMON block 'cxyzl' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/navier1.f:168.20: COMMON /SCRUZ/ TB1 (LX1,LY1,LZ1,LELV) 1 Warning: Named COMMON block 'scruz' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/makeq.f -o obj/makeq.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/navier0.f -o obj/navier0.o /scratch/user/fatihsinan/NEK/trunk/nek/navier1.f:1316.20: COMMON /CTMP1/ MFI (LX1,LY1,LZ1,LELV) 1 Warning: Named COMMON block 'ctmp1' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/navier2.f -o obj/navier2.o /scratch/user/fatihsinan/NEK/trunk/nek/navier1.f:2237.20: COMMON /SCRMG/ Y1 (LX1,LY1,LZ1,LELT) 1 Warning: Named COMMON block 'scrmg' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/navier3.f -o obj/navier3.o /scratch/user/fatihsinan/NEK/trunk/nek/navier1.f:3513.20: COMMON /CTMP0/ TA1 (LX1,LY1,LZ1,LELV) 1 Warning: Named COMMON block 'ctmp0' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/navier4.f -o obj/navier4.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/prepost.f -o obj/prepost.o /scratch/user/fatihsinan/NEK/trunk/nek/navier4.f:209.21: COMMON /ORTHOX/ Pbar(LTOT2),Pnew(LTOT2),Pbrr(ltot2) 1 Warning: Named COMMON block 'orthox' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/navier1.f:4206.20: COMMON /CTMP1/ RKX1 (LX1,LY1,LZ1,LELV) 1 Warning: Named COMMON block 'ctmp1' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/speclib.f -o obj/speclib.o /scratch/user/fatihsinan/NEK/trunk/nek/prepost.f:814.20: common /scruz/ ta1(lx1,ly1,lz1,lelv) 1 Warning: Named COMMON block 'scruz' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/navier4.f:1127.11: ivar(2) = m ! Update number of saved vectors 1 Warning: Array reference at (1) is out of bounds (2 > 1) in dimension 1 /scratch/user/fatihsinan/NEK/trunk/nek/prepost.f:2264.20: common /SCRNS/ u4(2+lxo*lxo*lxo*6*lelt) 1 Warning: Named COMMON block 'scrns' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/map2.f -o obj/map2.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/turb.f -o obj/turb.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/mvmesh.f -o obj/mvmesh.o /scratch/user/fatihsinan/NEK/trunk/nek/turb.f:578.20: COMMON /SCRUZ/ V1(LX1,LY1,LZ1,LELV) 1 Warning: Named COMMON block 'scruz' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/ic.f -o obj/ic.o /scratch/user/fatihsinan/NEK/trunk/nek/ic.f:966.38: $ call copy(t(1,1,1,1,2),sdmp2(1,1),ntott) 1 Warning: Array reference at (1) is out of bounds (2 > 1) in dimension 5 mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/ssolv.f -o obj/ssolv.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/planx.f -o obj/planx.o mpif77 -c -O3 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I./ /scratch/user/fatihsinan/NEK/trunk/nek/math.f -o obj/math.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/mxm_wrapper.f -o obj/mxm_wrapper.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/hmholtz.f -o obj/hmholtz.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/gfdm_par.f -o obj/gfdm_par.o /scratch/user/fatihsinan/NEK/trunk/nek/hmholtz.f:638.22: COMMON /CPRINT/ IFPRINT, IFHZPC 1 Warning: Named COMMON block 'cprint' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/gfdm_op.f -o obj/gfdm_op.o /scratch/user/fatihsinan/NEK/trunk/nek/hmholtz.f:2141.20: common /ctmp0/ ugunh(2*lx1*lz1*2*ldim*lelt) 1 Warning: Named COMMON block 'ctmp0' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/hmholtz.f:2143.20: common /ctmp1/ ur(lt),us(lt),ut(lt) 1 Warning: Named COMMON block 'ctmp1' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/gfdm_solve.f -o obj/gfdm_solve.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/subs1.f -o obj/subs1.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/subs2.f -o obj/subs2.o /scratch/user/fatihsinan/NEK/trunk/nek/subs1.f:1578.20: common /ctmp1/ exx(lx1*ly1*lz1*lelt) 1 Warning: Named COMMON block 'ctmp1' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/subs2.f:736.20: common /screv/ hfmask(lx1,lz1,6,lelt) 1 Warning: Named COMMON block 'screv' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/genbox.f -o obj/genbox.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/gmres.f -o obj/gmres.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/hsmg.f -o obj/hsmg.o /scratch/user/fatihsinan/NEK/trunk/nek/hsmg.f:296.20: common /hsmgw/ work(0:lwk-1),work2(0:lwk-1) 1 Warning: Named COMMON block 'hsmgw' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/hsmg.f:2046.20: common /ctmp0/ ur(lxyz),us(lxyz),ut(lxyz) 1 Warning: Named COMMON block 'ctmp0' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/hsmg.f:2535.20: common /scrvh/ h1 (lx1,ly1,lz1,lelv) 1 Warning: Named COMMON block 'scrvh' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/hsmg.f:2924.20: common /ctmp0/ w(100000) 1 Warning: Named COMMON block 'ctmp0' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/convect.f -o obj/convect.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/induct.f -o obj/induct.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/perturb.f -o obj/perturb.o /scratch/user/fatihsinan/NEK/trunk/nek/induct.f:189.20: COMMON /SCRNS/ TA1(LX1,LY1,LZ1,LELV) 1 Warning: Named COMMON block 'scrns' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/convect.f:946.20: common /ctmp1/ work(lxyz1,lelt) 1 Warning: Named COMMON block 'ctmp1' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/perturb.f:454.20: COMMON /SCRNS/ TA1(LX1,LY1,LZ1,LELV) 1 Warning: Named COMMON block 'scrns' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/convect.f:1538.20: common /scrdg/ uf(lf),uxf(lf),uyf(lf),uzf(lf),upwind_wgt(lf) 1 Warning: Named COMMON block 'scrdg' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/navier5.f -o obj/navier5.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/navier6.f -o obj/navier6.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/navier7.f -o obj/navier7.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/navier8.f -o obj/navier8.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/fast3d.f -o obj/fast3d.o /scratch/user/fatihsinan/NEK/trunk/nek/navier8.f:1449.20: common /ctmp0/ w(2,2,lx1),v(2,ly1,lz1,lelt) 1 Warning: Named COMMON block 'ctmp0' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/navier8.f:2022.21: common /scrns/ ind,etuple 1 Warning: Named COMMON block 'scrns' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/fasts.f -o obj/fasts.o /scratch/user/fatihsinan/NEK/trunk/nek/navier5.f:4085.20: common /ctmp1/ w(lt,lelt),ur(lt),us(lt),ut(lt),w1(2*lt) 1 Warning: Named COMMON block 'ctmp1' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/calcz.f -o obj/calcz.o mpicc -c -O2 -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -DPREFIX=jl_ /scratch/user/fatihsinan/NEK/trunk/nek/byte.c -o obj/byte.o mpicc -c -O2 -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -DPREFIX=jl_ /scratch/user/fatihsinan/NEK/trunk/nek/chelpers.c -o obj/chelpers.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/byte_mpi.f -o obj/byte_mpi.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/postpro.f -o obj/postpro.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cvode_driver.f -o obj/cvode_driver.o mpicc -c -O2 -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -DPREFIX=jl_ /scratch/user/fatihsinan/NEK/trunk/nek/nek_comm.c -o obj/nek_comm.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/init_plugin.f -o obj/init_plugin.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/setprop.f -o obj/setprop.o /scratch/user/fatihsinan/NEK/trunk/nek/postpro.f:1056.20: common /scrns/ vcurve(5,12,lblock),wk(5*12*lblock) 1 Warning: Named COMMON block 'scrns' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/postpro.f:1057.20: common /scruz/ icurve(12,lblock) 1 Warning: Named COMMON block 'scruz' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/qthermal.f -o obj/qthermal.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cvode_aux.f -o obj/cvode_aux.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/makeq_aux.f -o obj/makeq_aux.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/papi.f -o obj/papi.o /scratch/user/fatihsinan/NEK/trunk/nek/postpro.f:2111.20: common /scrns/ xyz(ldim,lt2) 1 Warning: Named COMMON block 'scrns' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/postpro.f:2112.20: common /scruz/ mid(lt2) ! Target proc id 1 Warning: Named COMMON block 'scruz' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/nek_in_situ.f -o obj/nek_in_situ.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cmt/drive1_cmt.f -o obj/drive1_cmt.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cmt/drive2_cmt.f -o obj/drive2_cmt.o /scratch/user/fatihsinan/NEK/trunk/nek/cmt/drive1_cmt.f:240.30: call copy(U(1,1,1,2,e),vx(1,1,1,e),nxyz1) 1 Warning: Array reference at (1) is out of bounds (2 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/drive1_cmt.f:241.30: call copy(U(1,1,1,3,e),vy(1,1,1,e),nxyz1) 1 Warning: Array reference at (1) is out of bounds (3 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/drive1_cmt.f:242.30: call copy(U(1,1,1,4,e),vz(1,1,1,e),nxyz1) 1 Warning: Array reference at (1) is out of bounds (4 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/drive1_cmt.f:243.30: call copy(U(1,1,1,5,e),t(1,1,1,e,3),nxyz1) 1 Warning: Array reference at (1) is out of bounds (5 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/drive1_cmt.f:243.45: call copy(U(1,1,1,5,e),t(1,1,1,e,3),nxyz1) 1 Warning: Array reference at (1) is out of bounds (3 > 1) in dimension 5 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/drive1_cmt.f:244.45: call copy(U(1,1,1,1,e),t(1,1,1,e,2),nxyz1) 1 Warning: Array reference at (1) is out of bounds (2 > 1) in dimension 5 mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cmt/driver3_cmt.f -o obj/driver3_cmt.o /scratch/user/fatihsinan/NEK/trunk/nek/cmt/driver3_cmt.f:16.42: call invcol3(vx(1,1,1,e),u(1,1,1,irpu,e),u(1,1,1,irg,e),nxyz) 1 Warning: Array reference at (1) is out of bounds (2 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/driver3_cmt.f:17.42: call invcol3(vy(1,1,1,e),u(1,1,1,irpv,e),u(1,1,1,irg,e),nxyz) 1 Warning: Array reference at (1) is out of bounds (3 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/driver3_cmt.f:19.42: call invcol3(vz(1,1,1,e),u(1,1,1,irpw,e),u(1,1,1,irg,e),nxyz) 1 Warning: Array reference at (1) is out of bounds (4 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/driver3_cmt.f:23.27: > u(1,1,1,irpu,e),u(1,1,1,irpv,e),u(1,1,1,irpw,e), 1 Warning: Array reference at (1) is out of bounds (2 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/driver3_cmt.f:23.43: > u(1,1,1,irpu,e),u(1,1,1,irpv,e),u(1,1,1,irpw,e), 1 Warning: Array reference at (1) is out of bounds (3 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/driver3_cmt.f:23.59: > u(1,1,1,irpu,e),u(1,1,1,irpv,e),u(1,1,1,irpw,e), 1 Warning: Array reference at (1) is out of bounds (4 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/driver3_cmt.f:24.27: > u(1,1,1,irpu,e),u(1,1,1,irpv,e),u(1,1,1,irpw,e),nxyz) 1 Warning: Array reference at (1) is out of bounds (2 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/driver3_cmt.f:24.43: > u(1,1,1,irpu,e),u(1,1,1,irpv,e),u(1,1,1,irpw,e),nxyz) 1 Warning: Array reference at (1) is out of bounds (3 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/driver3_cmt.f:24.59: > u(1,1,1,irpu,e),u(1,1,1,irpv,e),u(1,1,1,irpw,e),nxyz) 1 Warning: Array reference at (1) is out of bounds (4 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/driver3_cmt.f:26.35: call vdot2(scr,u(1,1,1,irpu,e),u(1,1,1,irpv,e), 1 Warning: Array reference at (1) is out of bounds (2 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/driver3_cmt.f:26.51: call vdot2(scr,u(1,1,1,irpu,e),u(1,1,1,irpv,e), 1 Warning: Array reference at (1) is out of bounds (3 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/driver3_cmt.f:27.35: > u(1,1,1,irpu,e),u(1,1,1,irpv,e),nxyz) 1 Warning: Array reference at (1) is out of bounds (2 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/driver3_cmt.f:27.51: > u(1,1,1,irpu,e),u(1,1,1,irpv,e),nxyz) 1 Warning: Array reference at (1) is out of bounds (3 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/driver3_cmt.f:64.24: vtrans(i,j,k,e,icv)= cv*rho 1 Warning: Array reference at (1) is out of bounds (3 > 2) in dimension 5 mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cmt/face.f -o obj/face.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cmt/ausm.f -o obj/ausm.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cmt/MixtPerf.f -o obj/MixtPerf.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cmt/intpdiff.f -o obj/intpdiff.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cmt/diffusive_cmt.f -o obj/diffusive_cmt.o /scratch/user/fatihsinan/NEK/trunk/nek/cmt/diffusive_cmt.f:305.26: vdiff(i,1,1,e,ilam) = lambda 1 Warning: Array reference at (1) is out of bounds (3 > 2) in dimension 5 mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cmt/eqnsolver_cmt.f -o obj/eqnsolver_cmt.o /scratch/user/fatihsinan/NEK/trunk/nek/cmt/eqnsolver_cmt.f:149.37: > vdiff(1,1,1,e,ilam), 1 Warning: Array reference at (1) is out of bounds (3 > 2) in dimension 5 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/eqnsolver_cmt.f:151.38: > vtrans(1,1,1,e,icv)) 1 Warning: Array reference at (1) is out of bounds (3 > 2) in dimension 5 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/eqnsolver_cmt.f:209.20: common /ctmp1/ ur(ldd),us(ldd),ut(ldd),ju(ldd),ud(ldd),tu(ldd) 1 Warning: Named COMMON block 'ctmp1' at (1) shall be of the same size mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cmt/surface_fluxes.f -o obj/surface_fluxes.o /scratch/user/fatihsinan/NEK/trunk/nek/cmt/eqnsolver_cmt.f:261.20: common /ctmp1/ ur(ldd),us(ldd),ut(ldd),ju(ldd),ud(ldd),tu(ldd) 1 Warning: Named COMMON block 'ctmp1' at (1) shall be of the same size /scratch/user/fatihsinan/NEK/trunk/nek/cmt/eqnsolver_cmt.f:331.40: & + U(i,j,k,4,e)*FFZ)/ U(i,j,k,1,e) 1 Warning: Array reference at (1) is out of bounds (4 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/eqnsolver_cmt.f:330.59: usrf(i,j,k,5) = (U(i,j,k,2,e)*FFX + U(i,j,k,3,e)*FFY 1 Warning: Array reference at (1) is out of bounds (3 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/eqnsolver_cmt.f:330.40: usrf(i,j,k,5) = (U(i,j,k,2,e)*FFX + U(i,j,k,3,e)*FFY 1 Warning: Array reference at (1) is out of bounds (2 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/surface_fluxes.f:41.37: call fillq(icvf,vtrans(1,1,1,1,icv),fatface(iqm),fatface(iqp)) 1 Warning: Array reference at (1) is out of bounds (3 > 2) in dimension 5 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/surface_fluxes.f:46.40: call fillq(ilamf,vdiff(1,1,1,1,ilam),fatface(iqm),fatface(iqp)) 1 Warning: Array reference at (1) is out of bounds (3 > 2) in dimension 5 mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cmt/bc.f -o obj/bc.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cmt/inflow_bc.f -o obj/inflow_bc.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cmt/outflow_bc.f -o obj/outflow_bc.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cmt/wall_bc.f -o obj/wall_bc.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cmt/step.f -o obj/step.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cmt/filters_cmt.f -o obj/filters_cmt.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ /scratch/user/fatihsinan/NEK/trunk/nek/cmt/diagnostics.f -o obj/diagnostics.o mpif77 -c -O3 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I./ /scratch/user/fatihsinan/NEK/trunk/nek/mxm_std.f -o obj/mxm_std.o /scratch/user/fatihsinan/NEK/trunk/nek/cmt/diagnostics.f:80.44: call copy(otvar(1,1,1,e,2),u(1,1,1,2,e),n) 1 Warning: Array reference at (1) is out of bounds (2 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/diagnostics.f:81.44: call copy(otvar(1,1,1,e,3),u(1,1,1,3,e),n) 1 Warning: Array reference at (1) is out of bounds (3 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/diagnostics.f:82.44: call copy(otvar(1,1,1,e,4),u(1,1,1,4,e),n) 1 Warning: Array reference at (1) is out of bounds (4 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/diagnostics.f:83.44: call copy(otvar(1,1,1,e,5),u(1,1,1,5,e),n) 1 Warning: Array reference at (1) is out of bounds (5 > 1) in dimension 4 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/diagnostics.f:192.33: $ ,rhseqs(1),rhseqs(2),rhseqs(3),rhseqs(4) 1 Warning: Array reference at (1) is out of bounds (2 > 1) in dimension 1 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/diagnostics.f:192.43: $ ,rhseqs(1),rhseqs(2),rhseqs(3),rhseqs(4) 1 Warning: Array reference at (1) is out of bounds (3 > 1) in dimension 1 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/diagnostics.f:192.53: $ ,rhseqs(1),rhseqs(2),rhseqs(3),rhseqs(4) 1 Warning: Array reference at (1) is out of bounds (4 > 1) in dimension 1 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/diagnostics.f:193.23: $ ,rhseqs(5) 1 Warning: Array reference at (1) is out of bounds (5 > 1) in dimension 1 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/diagnostics.f:205.33: $ ,rhseqs(1),rhseqs(2),rhseqs(3),rhseqs(4) 1 Warning: Array reference at (1) is out of bounds (2 > 1) in dimension 1 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/diagnostics.f:205.43: $ ,rhseqs(1),rhseqs(2),rhseqs(3),rhseqs(4) 1 Warning: Array reference at (1) is out of bounds (3 > 1) in dimension 1 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/diagnostics.f:205.53: $ ,rhseqs(1),rhseqs(2),rhseqs(3),rhseqs(4) 1 Warning: Array reference at (1) is out of bounds (4 > 1) in dimension 1 /scratch/user/fatihsinan/NEK/trunk/nek/cmt/diagnostics.f:206.23: $ ,rhseqs(5) 1 Warning: Array reference at (1) is out of bounds (5 > 1) in dimension 1 mpif77 -c -O0 -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I./ /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/blas.f -o obj/blas.o mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ -DUNORDERED_MAP_NS=std::tr1 -DHAVE_UNORDERED_MAP=tr1/unordered_map -DHAVE_UNORDERED_SET=tr1/unordered_set -I/scratch/user/fatihsinan/MOAB/hdf5/hdf5-1.8.16/gcc/include -isystem /scratch/user/fatihsinan/MOAB/hdf5/hdf5-1.8.16/gcc/include -DTEMPLATE_SPECIALIZATION -DTEMPLATE_FUNC_SPECIALIZATION -DHAVE_VSNPRINTF -D_FILE_OFFSET_BITS=64 -DUSE_MPI -DHDF5_FILE -DHDF5_PARALLEL -DHAVE_ZOLTAN -I/scratch/user/fatihsinan/MOAB/moab_nek/trunk-test/gcc/include /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f -o obj/moab.o SIZE:134.22: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/MOABCORE:21: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/NEKMOAB:53: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:79: parameter(lelcmt = 1 ) ! # of cmt elements 1 Error: Symbol 'lelcmt' at (1) has no IMPLICIT type SIZE:134.22: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/MOABCORE:21: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/NEKMOAB:53: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:121: parameter(lelcmt = 1 ) ! # of cmt elements 1 Error: Symbol 'lelcmt' at (1) has no IMPLICIT type /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:131.23: pointer (tagv_ptr, tag_vals(1)) 1 Warning: Cray pointer at (1) has 4 bytes of precision; memory addresses require 8 bytes /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:132.22: pointer (adj_ptr, adjs(1)) 1 Warning: Cray pointer at (1) has 4 bytes of precision; memory addresses require 8 bytes /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:133.24: pointer (verts_ptr, verts(1)) 1 Warning: Cray pointer at (1) has 4 bytes of precision; memory addresses require 8 bytes /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:134.24: pointer (elems_ptr, elems(1)) 1 Warning: Cray pointer at (1) has 4 bytes of precision; memory addresses require 8 bytes /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:135.25: pointer (offset_ptr, offset(1)) 1 Warning: Cray pointer at (1) has 4 bytes of precision; memory addresses require 8 bytes SIZE:134.22: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/MOABCORE:21: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/NEKMOAB:53: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:218: parameter(lelcmt = 1 ) ! # of cmt elements 1 Error: Symbol 'lelcmt' at (1) has no IMPLICIT type GEOM:62.26: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:219: $ ,etalph(lx1*lz1,2*ldim,lelt) 1 Error: Symbol 'etalph' at (1) has no IMPLICIT type PARALLEL:35.45: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:218: common /xcdg_ints/ dg_hndlx,ndg_facex 1 Error: Symbol 'ndg_facex' at (1) has no IMPLICIT type GEOM:49.23: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:219: COMMON /GSURF/unr (lx1*lz1,6,lelt) 1 Error: Symbol 'unr' at (1) has no IMPLICIT type GEOM:50.23: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:219: $ ,uns (lx1*lz1,6,lelt) 1 Error: Symbol 'uns' at (1) has no IMPLICIT type GEOM:51.23: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:219: $ ,unt (lx1*lz1,6,lelt) 1 Error: Symbol 'unt' at (1) has no IMPLICIT type SIZE:134.22: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/MOABCORE:21: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/NEKMOAB:53: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:265: parameter(lelcmt = 1 ) ! # of cmt elements 1 Error: Symbol 'lelcmt' at (1) has no IMPLICIT type SIZE:134.22: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/MOABCORE:21: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/NEKMOAB:53: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:281: parameter(lelcmt = 1 ) ! # of cmt elements 1 Error: Symbol 'lelcmt' at (1) has no IMPLICIT type /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/blas.f:9764.15: GO TO IGO,(120,150,180,210) 1 Warning: Deleted feature: Assigned GOTO statement at (1) /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/blas.f:9770.72: ASSIGN 120 TO IGO 1 Warning: Deleted feature: ASSIGN statement at (1) /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/blas.f:9782.72: ASSIGN 150 TO IGO 1 Warning: Deleted feature: ASSIGN statement at (1) /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/blas.f:9795.72: ASSIGN 180 TO IGO 1 Warning: Deleted feature: ASSIGN statement at (1) /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/blas.f:9806.72: ASSIGN 210 TO IGO 1 Warning: Deleted feature: ASSIGN statement at (1) SIZE:134.22: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/MOABCORE:21: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/NEKMOAB:53: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:414: parameter(lelcmt = 1 ) ! # of cmt elements 1 Error: Symbol 'lelcmt' at (1) has no IMPLICIT type SIZE:134.22: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/MOABCORE:21: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/NEKMOAB:53: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:456: parameter(lelcmt = 1 ) ! # of cmt elements 1 Error: Symbol 'lelcmt' at (1) has no IMPLICIT type SIZE:134.22: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/MOABCORE:21: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/NEKMOAB:53: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:497: parameter(lelcmt = 1 ) ! # of cmt elements 1 Error: Symbol 'lelcmt' at (1) has no IMPLICIT type PARALLEL:35.45: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:497: common /xcdg_ints/ dg_hndlx,ndg_facex 1 Error: Symbol 'ndg_facex' at (1) has no IMPLICIT type mpif77 -c -O2 -fdefault-real-8 -fdefault-double-8 -x f77-cpp-input -fcray-pointer -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG -I/scratch/user/fatihsinan/NEK/examples/moab -I/scratch/user/fatihsinan/NEK/trunk/nek -I/scratch/user/fatihsinan/NEK/trunk/nek/cmt -I./ -DUNORDERED_MAP_NS=std::tr1 -DHAVE_UNORDERED_MAP=tr1/unordered_map -DHAVE_UNORDERED_SET=tr1/unordered_set -I/scratch/user/fatihsinan/MOAB/hdf5/hdf5-1.8.16/gcc/include -isystem /scratch/user/fatihsinan/MOAB/hdf5/hdf5-1.8.16/gcc/include -DTEMPLATE_SPECIALIZATION -DTEMPLATE_FUNC_SPECIALIZATION -DHAVE_VSNPRINTF -D_FILE_OFFSET_BITS=64 -DUSE_MPI -DHDF5_FILE -DHDF5_PARALLEL -DHAVE_ZOLTAN -I/scratch/user/fatihsinan/MOAB/moab_nek/trunk-test/gcc/include /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/imeshutil.f -o obj/imeshutil.o SIZE:134.22: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:557: parameter(lelcmt = 1 ) ! # of cmt elements 1 Error: Symbol 'lelcmt' at (1) has no IMPLICIT type /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:564.21: pointer(tag_ptr, nekid(1)) 1 Warning: Cray pointer at (1) has 4 bytes of precision; memory addresses require 8 bytes SIZE:134.22: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:732: parameter(lelcmt = 1 ) ! # of cmt elements 1 Error: Symbol 'lelcmt' at (1) has no IMPLICIT type /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:820.21: pointer(tag_ptr, gid(1)) 1 Warning: Cray pointer at (1) has 4 bytes of precision; memory addresses require 8 bytes SIZE:134.22: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/MOABCORE:21: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/NEKMOAB:53: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:823: parameter(lelcmt = 1 ) ! # of cmt elements 1 Error: Symbol 'lelcmt' at (1) has no IMPLICIT type PARALLEL:35.45: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:823: common /xcdg_ints/ dg_hndlx,ndg_facex 1 Error: Symbol 'ndg_facex' at (1) has no IMPLICIT type SIZE:134.22: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/MOABCORE:21: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/NEKMOAB:53: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:864: parameter(lelcmt = 1 ) ! # of cmt elements 1 Error: Symbol 'lelcmt' at (1) has no IMPLICIT type /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:874.26: pointer (connect_ptr, connect(0:1)) 1 Warning: Cray pointer at (1) has 4 bytes of precision; memory addresses require 8 bytes SIZE:134.22: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/MOABCORE:21: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/NEKMOAB:53: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:924: parameter(lelcmt = 1 ) ! # of cmt elements 1 Error: Symbol 'lelcmt' at (1) has no IMPLICIT type /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:927.25: pointer(connect_ptr, connect(0:1)) 1 Warning: Cray pointer at (1) has 4 bytes of precision; memory addresses require 8 bytes SIZE:134.22: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/MOABCORE:21: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/NEKMOAB:53: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:971: parameter(lelcmt = 1 ) ! # of cmt elements 1 Error: Symbol 'lelcmt' at (1) has no IMPLICIT type SIZE:134.22: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/MOABCORE:21: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/NEKMOAB:53: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:1033: parameter(lelcmt = 1 ) ! # of cmt elements 1 Error: Symbol 'lelcmt' at (1) has no IMPLICIT type mpicc -c -O2 -DMOAB -DMPI -DUNDERSCORE -DGLOBAL_LONG_LONG /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/imeshcutil.c -o obj/imeshcutil.o SIZE:134.22: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/MOABCORE:21: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/NEKMOAB:53: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:1158: parameter(lelcmt = 1 ) ! # of cmt elements 1 Error: Symbol 'lelcmt' at (1) has no IMPLICIT type PARALLEL:35.45: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:1158: common /xcdg_ints/ dg_hndlx,ndg_facex 1 Error: Symbol 'ndg_facex' at (1) has no IMPLICIT type SIZE:134.22: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/MOABCORE:21: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/NEKMOAB:53: Included at /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/moab.f:1174: parameter(lelcmt = 1 ) ! # of cmt elements 1 Error: Symbol 'lelcmt' at (1) has no IMPLICIT type Fatal Error: Error count reached limit of 25. make: *** [obj/moab.o] Error 1 make: *** Waiting for unfinished jobs.... /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/blas.f:17277.15: GO TO IGO,(120,150,180,210) 1 Warning: Deleted feature: Assigned GOTO statement at (1) /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/blas.f:17283.72: ASSIGN 120 TO IGO 1 Warning: Deleted feature: ASSIGN statement at (1) /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/blas.f:17295.72: ASSIGN 150 TO IGO 1 Warning: Deleted feature: ASSIGN statement at (1) /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/blas.f:17308.72: ASSIGN 180 TO IGO 1 Warning: Deleted feature: ASSIGN statement at (1) /scratch/user/fatihsinan/NEK/trunk/nek/3rd_party/blas.f:17319.72: ASSIGN 210 TO IGO 1 Warning: Deleted feature: ASSIGN statement at (1) From nek5000-users at lists.mcs.anl.gov Thu Nov 19 15:37:35 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 19 Nov 2015 21:37:35 +0000 Subject: [Nek5000-users] NEK-MOAB coupling In-Reply-To: References: Message-ID: Yes, thanks, Fatih (and Emilian), we are working to resolve the issue but for now you could use an earlier version of the (experimental!) Nek-MOAB interface, e.g. from revision 1061. Aleks ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Thursday, November 19, 2015 3:33 PM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] NEK-MOAB coupling Dear Neks, I think recent version of Nek has some problems to couple Nek with MOAB, I coupled Nek and MOAB before. But with current version of Nek (1090) , it can't compile with MOAB. I attached the makenek output. ( P.S. MOAB version 4.6.3 and compiled with mpi and parallel hdf5 and passed all the tests) Thanks, Fatih -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu Nov 19 16:14:15 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 19 Nov 2015 16:14:15 -0600 Subject: [Nek5000-users] NEK-MOAB coupling In-Reply-To: References: Message-ID: Thank you LAleks , I just checked out 1061 and run examples/moab as a test, it is working !!! Fatih On Thu, Nov 19, 2015 at 3:37 PM, wrote: > Yes, thanks, Fatih (and Emilian), > > we are working to resolve the issue but for now you could use an earlier > version of the (experimental!) Nek-MOAB interface, e.g. from revision 1061. > > Aleks > > ------------------------------ > *From:* nek5000-users-bounces at lists.mcs.anl.gov [ > nek5000-users-bounces at lists.mcs.anl.gov] on behalf of > nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] > *Sent:* Thursday, November 19, 2015 3:33 PM > *To:* nek5000-users at lists.mcs.anl.gov > *Subject:* [Nek5000-users] NEK-MOAB coupling > > Dear Neks, > > I think recent version of Nek has some problems to couple Nek with MOAB, I > coupled Nek and MOAB before. But with current version of Nek (1090) , it > can't compile with MOAB. I attached the makenek output. > > ( P.S. MOAB version 4.6.3 and compiled with mpi and parallel hdf5 and > passed all the tests) > > Thanks, > > Fatih > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > -- Fatih Sinan Sarikurt Graduate Research Assistant Texas A&M University Department of Nuclear Engineering fatihsinan at tamu.edu Mobile: +1 979 739 42 93 -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Fri Nov 20 07:46:22 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 20 Nov 2015 14:46:22 +0100 Subject: [Nek5000-users] low order interpolation Message-ID: Dear neks I don't know how Nek5000 performs the interpolation with hpts() routine but I have some problem when I use this last. Using Matlab I built an hpts.in file that contains the points of a structured and not constant mesh, then using Nek in postprocessing mode I evaluate the values of velocity an pressure in this points. When I reshape the hpts.out in order to visualize my domain, there are many zones that have no-sense values!. I performed the same interpolation in tecplot and it works fine. my question is: there's any way to perform a first order interpolation in Nek5000? Sincerely Alessandro -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Fri Nov 20 10:29:58 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 20 Nov 2015 09:29:58 -0700 Subject: [Nek5000-users] low order interpolation In-Reply-To: References: Message-ID: Hi Alessandro, I use exactly the same procedure for my case. Just wondering, by any chance, are there any points in hpts.in which are outside bounds of the original domain, or you may resize the domain in userdat(), while postprocessing in nek5000. Best Regards, Tanmoy On Fri, Nov 20, 2015 at 6:46 AM, wrote: > Dear neks > > I don't know how Nek5000 performs the interpolation with hpts() routine > but I have some problem when I use this last. Using Matlab I built an > hpts.in file that contains the points of a structured and not constant > mesh, then using Nek in postprocessing mode I evaluate the values of > velocity an pressure in this points. When I reshape the hpts.out in order > to visualize my domain, there are many zones that have no-sense values!. I > performed the same interpolation in tecplot and it works fine. > > my question is: there's any way to perform a first order interpolation in > Nek5000? > > Sincerely > Alessandro > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Fri Nov 20 11:41:20 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 20 Nov 2015 18:41:20 +0100 Subject: [Nek5000-users] low order interpolation In-Reply-To: References: Message-ID: Hi Tanmoy thanks for the quick answer, unfortunately I didn't commit this kinds of errors. I have already used my matlab routine for other cases and it worked well, so I'm pretty sure that there isn't errors. The only differences is that I'm using this routine to interpolate the hpts.in points on more complex grid and on a turbulent field. It seams that I have the Runge's phenomenon. Is this possible with the hpts routine? Sincerely Alessandro 2015-11-20 17:29 GMT+01:00 : > Hi Alessandro, > > I use exactly the same procedure for my case. Just wondering, by any > chance, are there any points in hpts.in which are outside bounds of the > original domain, or you may resize the domain in userdat(), while > postprocessing in nek5000. > > Best Regards, > Tanmoy > > On Fri, Nov 20, 2015 at 6:46 AM, wrote: > >> Dear neks >> >> I don't know how Nek5000 performs the interpolation with hpts() routine >> but I have some problem when I use this last. Using Matlab I built an >> hpts.in file that contains the points of a structured and not constant >> mesh, then using Nek in postprocessing mode I evaluate the values of >> velocity an pressure in this points. When I reshape the hpts.out in order >> to visualize my domain, there are many zones that have no-sense values!. I >> performed the same interpolation in tecplot and it works fine. >> >> my question is: there's any way to perform a first order interpolation in >> Nek5000? >> >> Sincerely >> Alessandro >> >> _______________________________________________ >> Nek5000-users mailing list >> Nek5000-users at lists.mcs.anl.gov >> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users >> >> > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Sat Nov 21 16:53:59 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sat, 21 Nov 2015 15:53:59 -0700 Subject: [Nek5000-users] low order interpolation In-Reply-To: References: Message-ID: Hi Alessandro, Ok. But, I was not talking about your matlab routines, but rather I was talking about your nek5000 postprocessing mode. I guess they are correct, since you got good results with Tecplot. But anyway, the bottom line is, nek5000 uses spectral interpolation routine, so even the higher order polynomials are orthogonal, so it is unlikely that you will have Runge's phenomenon. Also, I myself use hpts routines for interpolating a turbulent field Best Regards, Tanmoy On Fri, Nov 20, 2015 at 10:41 AM, wrote: > Hi Tanmoy > > thanks for the quick answer, unfortunately I didn't commit this kinds of > errors. I have already used my matlab routine for other cases and it worked > well, so I'm pretty sure that there isn't errors. The only differences is > that I'm using this routine to interpolate the hpts.in points on more > complex grid and on a turbulent field. It seams that I have the Runge's > phenomenon. Is this possible with the hpts routine? > > Sincerely > Alessandro > > > > > > 2015-11-20 17:29 GMT+01:00 : > >> Hi Alessandro, >> >> I use exactly the same procedure for my case. Just wondering, by any >> chance, are there any points in hpts.in which are outside bounds of the >> original domain, or you may resize the domain in userdat(), while >> postprocessing in nek5000. >> >> Best Regards, >> Tanmoy >> >> On Fri, Nov 20, 2015 at 6:46 AM, wrote: >> >>> Dear neks >>> >>> I don't know how Nek5000 performs the interpolation with hpts() routine >>> but I have some problem when I use this last. Using Matlab I built an >>> hpts.in file that contains the points of a structured and not constant >>> mesh, then using Nek in postprocessing mode I evaluate the values of >>> velocity an pressure in this points. When I reshape the hpts.out in order >>> to visualize my domain, there are many zones that have no-sense values!. I >>> performed the same interpolation in tecplot and it works fine. >>> >>> my question is: there's any way to perform a first order interpolation >>> in Nek5000? >>> >>> Sincerely >>> Alessandro >>> >>> _______________________________________________ >>> Nek5000-users mailing list >>> Nek5000-users at lists.mcs.anl.gov >>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users >>> >>> >> >> _______________________________________________ >> Nek5000-users mailing list >> Nek5000-users at lists.mcs.anl.gov >> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users >> >> > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Sun Nov 22 12:44:10 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sun, 22 Nov 2015 19:44:10 +0100 Subject: [Nek5000-users] low order interpolation In-Reply-To: References: Message-ID: Hi Tanmoy I really don't understand why the interpolation fails.. I hope that someone can help me because a sample low order interpolation could solve my problem. Thanks anyway Alessandro 2015-11-21 23:53 GMT+01:00 : > Hi Alessandro, > > Ok. But, I was not talking about your matlab routines, but rather I was > talking about your nek5000 postprocessing mode. I guess they are correct, > since you got good results with Tecplot. > > But anyway, the bottom line is, nek5000 uses spectral interpolation > routine, so even the higher order polynomials are orthogonal, so it is > unlikely that you will have Runge's phenomenon. > > Also, I myself use hpts routines for interpolating a turbulent field > > Best Regards, > Tanmoy > > On Fri, Nov 20, 2015 at 10:41 AM, wrote: > >> Hi Tanmoy >> >> thanks for the quick answer, unfortunately I didn't commit this kinds of >> errors. I have already used my matlab routine for other cases and it worked >> well, so I'm pretty sure that there isn't errors. The only differences is >> that I'm using this routine to interpolate the hpts.in points on more >> complex grid and on a turbulent field. It seams that I have the Runge's >> phenomenon. Is this possible with the hpts routine? >> >> Sincerely >> Alessandro >> >> >> >> >> >> 2015-11-20 17:29 GMT+01:00 : >> >>> Hi Alessandro, >>> >>> I use exactly the same procedure for my case. Just wondering, by any >>> chance, are there any points in hpts.in which are outside bounds of the >>> original domain, or you may resize the domain in userdat(), while >>> postprocessing in nek5000. >>> >>> Best Regards, >>> Tanmoy >>> >>> On Fri, Nov 20, 2015 at 6:46 AM, >>> wrote: >>> >>>> Dear neks >>>> >>>> I don't know how Nek5000 performs the interpolation with hpts() routine >>>> but I have some problem when I use this last. Using Matlab I built an >>>> hpts.in file that contains the points of a structured and not constant >>>> mesh, then using Nek in postprocessing mode I evaluate the values of >>>> velocity an pressure in this points. When I reshape the hpts.out in order >>>> to visualize my domain, there are many zones that have no-sense values!. I >>>> performed the same interpolation in tecplot and it works fine. >>>> >>>> my question is: there's any way to perform a first order interpolation >>>> in Nek5000? >>>> >>>> Sincerely >>>> Alessandro >>>> >>>> _______________________________________________ >>>> Nek5000-users mailing list >>>> Nek5000-users at lists.mcs.anl.gov >>>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users >>>> >>>> >>> >>> _______________________________________________ >>> Nek5000-users mailing list >>> Nek5000-users at lists.mcs.anl.gov >>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users >>> >>> >> >> _______________________________________________ >> Nek5000-users mailing list >> Nek5000-users at lists.mcs.anl.gov >> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users >> >> > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Mon Nov 23 08:32:46 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 23 Nov 2015 15:32:46 +0100 Subject: [Nek5000-users] errors Message-ID: Hi Neks, can someone tell me what means these errors? 1 100 **ERROR**: Failed in HMHOLTZ: VELX 7.5730E+00 8.9997E+04 1.0000E-08 1.00000000000000002E-008 p22 1 1 1 1 Helmholtz VELY F: 1.1292E-11 1.0000E-08 1.0000E-03 1.0000E+02 1 Hmholtz VELY: 0 1.1292E-11 1.1292E-11 1.0000E-08 1.00000000000000002E-008 p22 1 1 1 1 Helmholtz VELZ F: 6.8078E-12 1.0000E-08 1.0000E-03 1.0000E+02 1 Hmholtz VELZ: 0 6.8078E-12 6.8078E-12 1.0000E-08 L1/L2 DIV(V) : 2.3607E-17 2.1252E-05 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.3607E-17 2.1252E-05 1 1.0000E-02 1.0853E+01 Fluid done and: ERROR: .sch file already exists. Use IFSCHCLOB=F to disable this check BUT BEWARE!!!!!! ierr= 1 They are came out at first step of my sim. Thanks From nek5000-users at lists.mcs.anl.gov Mon Nov 23 14:12:33 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 23 Nov 2015 20:12:33 +0000 Subject: [Nek5000-users] errors In-Reply-To: References: Message-ID: Hi, My guess is that you are not running the case with nek scripts -- see http://www.mcs.anl.gov/~obabko/nekrun.pdf The scripts like nekb or nekbmpi move/remove history file .sch at the start. Aleks ________________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Monday, November 23, 2015 8:32 AM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] errors Hi Neks, can someone tell me what means these errors? 1 100 **ERROR**: Failed in HMHOLTZ: VELX 7.5730E+00 8.9997E+04 1.0000E-08 1.00000000000000002E-008 p22 1 1 1 1 Helmholtz VELY F: 1.1292E-11 1.0000E-08 1.0000E-03 1.0000E+02 1 Hmholtz VELY: 0 1.1292E-11 1.1292E-11 1.0000E-08 1.00000000000000002E-008 p22 1 1 1 1 Helmholtz VELZ F: 6.8078E-12 1.0000E-08 1.0000E-03 1.0000E+02 1 Hmholtz VELZ: 0 6.8078E-12 6.8078E-12 1.0000E-08 L1/L2 DIV(V) : 2.3607E-17 2.1252E-05 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.3607E-17 2.1252E-05 1 1.0000E-02 1.0853E+01 Fluid done and: ERROR: .sch file already exists. Use IFSCHCLOB=F to disable this check BUT BEWARE!!!!!! ierr= 1 They are came out at first step of my sim. Thanks _______________________________________________ Nek5000-users mailing list Nek5000-users at lists.mcs.anl.gov https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users From nek5000-users at lists.mcs.anl.gov Mon Nov 23 17:13:21 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 23 Nov 2015 23:13:21 +0000 Subject: [Nek5000-users] errors In-Reply-To: References: Message-ID: Hi, The first error was because it did not converge to the level you set (parameter 12) when solving for x velocity. You may check your initial condition or other set up. As for the second error, you would have a *.sch file every time you run nek, but there is a check to see whether the file exists. You may just delete the *.sch file before you run nek again. Cheers, Tony ________________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov on behalf of nek5000-users-request at lists.mcs.anl.gov Sent: 23 November 2015 18:00 To: nek5000-users at lists.mcs.anl.gov Subject: Nek5000-users Digest, Vol 81, Issue 21 Send Nek5000-users mailing list submissions to nek5000-users at lists.mcs.anl.gov To subscribe or unsubscribe via the World Wide Web, visit https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users or, via email, send a message with subject or body 'help' to nek5000-users-request at lists.mcs.anl.gov You can reach the person managing the list at nek5000-users-owner at lists.mcs.anl.gov When replying, please edit your Subject line so it is more specific than "Re: Contents of Nek5000-users digest..." Today's Topics: 1. Re: low order interpolation (nek5000-users at lists.mcs.anl.gov) 2. errors (nek5000-users at lists.mcs.anl.gov) ---------------------------------------------------------------------- Message: 1 Date: Sun, 22 Nov 2015 19:44:10 +0100 From: nek5000-users at lists.mcs.anl.gov To: nek5000-users at lists.mcs.anl.gov Subject: Re: [Nek5000-users] low order interpolation Message-ID: Content-Type: text/plain; charset="utf-8" Hi Tanmoy I really don't understand why the interpolation fails.. I hope that someone can help me because a sample low order interpolation could solve my problem. Thanks anyway Alessandro 2015-11-21 23:53 GMT+01:00 : > Hi Alessandro, > > Ok. But, I was not talking about your matlab routines, but rather I was > talking about your nek5000 postprocessing mode. I guess they are correct, > since you got good results with Tecplot. > > But anyway, the bottom line is, nek5000 uses spectral interpolation > routine, so even the higher order polynomials are orthogonal, so it is > unlikely that you will have Runge's phenomenon. > > Also, I myself use hpts routines for interpolating a turbulent field > > Best Regards, > Tanmoy > > On Fri, Nov 20, 2015 at 10:41 AM, wrote: > >> Hi Tanmoy >> >> thanks for the quick answer, unfortunately I didn't commit this kinds of >> errors. I have already used my matlab routine for other cases and it worked >> well, so I'm pretty sure that there isn't errors. The only differences is >> that I'm using this routine to interpolate the hpts.in points on more >> complex grid and on a turbulent field. It seams that I have the Runge's >> phenomenon. Is this possible with the hpts routine? >> >> Sincerely >> Alessandro >> >> >> >> >> >> 2015-11-20 17:29 GMT+01:00 : >> >>> Hi Alessandro, >>> >>> I use exactly the same procedure for my case. Just wondering, by any >>> chance, are there any points in hpts.in which are outside bounds of the >>> original domain, or you may resize the domain in userdat(), while >>> postprocessing in nek5000. >>> >>> Best Regards, >>> Tanmoy >>> >>> On Fri, Nov 20, 2015 at 6:46 AM, >>> wrote: >>> >>>> Dear neks >>>> >>>> I don't know how Nek5000 performs the interpolation with hpts() routine >>>> but I have some problem when I use this last. Using Matlab I built an >>>> hpts.in file that contains the points of a structured and not constant >>>> mesh, then using Nek in postprocessing mode I evaluate the values of >>>> velocity an pressure in this points. When I reshape the hpts.out in order >>>> to visualize my domain, there are many zones that have no-sense values!. I >>>> performed the same interpolation in tecplot and it works fine. >>>> >>>> my question is: there's any way to perform a first order interpolation >>>> in Nek5000? >>>> >>>> Sincerely >>>> Alessandro >>>> >>>> _______________________________________________ >>>> Nek5000-users mailing list >>>> Nek5000-users at lists.mcs.anl.gov >>>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users >>>> >>>> >>> >>> _______________________________________________ >>> Nek5000-users mailing list >>> Nek5000-users at lists.mcs.anl.gov >>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users >>> >>> >> >> _______________________________________________ >> Nek5000-users mailing list >> Nek5000-users at lists.mcs.anl.gov >> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users >> >> > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Message: 2 Date: Mon, 23 Nov 2015 15:32:46 +0100 From: nek5000-users at lists.mcs.anl.gov To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] errors Message-ID: Content-Type: text/plain; charset=utf-8; format=flowed Hi Neks, can someone tell me what means these errors? 1 100 **ERROR**: Failed in HMHOLTZ: VELX 7.5730E+00 8.9997E+04 1.0000E-08 1.00000000000000002E-008 p22 1 1 1 1 Helmholtz VELY F: 1.1292E-11 1.0000E-08 1.0000E-03 1.0000E+02 1 Hmholtz VELY: 0 1.1292E-11 1.1292E-11 1.0000E-08 1.00000000000000002E-008 p22 1 1 1 1 Helmholtz VELZ F: 6.8078E-12 1.0000E-08 1.0000E-03 1.0000E+02 1 Hmholtz VELZ: 0 6.8078E-12 6.8078E-12 1.0000E-08 L1/L2 DIV(V) : 2.3607E-17 2.1252E-05 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.3607E-17 2.1252E-05 1 1.0000E-02 1.0853E+01 Fluid done and: ERROR: .sch file already exists. Use IFSCHCLOB=F to disable this check BUT BEWARE!!!!!! ierr= 1 They are came out at first step of my sim. Thanks ------------------------------ _______________________________________________ Nek5000-users mailing list Nek5000-users at lists.mcs.anl.gov https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users End of Nek5000-users Digest, Vol 81, Issue 21 ********************************************* From nek5000-users at lists.mcs.anl.gov Tue Nov 24 06:31:25 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 24 Nov 2015 13:31:25 +0100 Subject: [Nek5000-users] Coupled channel boundary conditions Message-ID: Dear All, I have a domain composed of two channels, one on top of the other. The top channel has periodic boundary conditions in the stream-wise direction. How can I impose the top channel boundary condition as the inlet boundary condition for the bottom channel? Or alternatively, how do I pass conditions in any plane in the top channel to the bottom channel inlet condition? To make things easier both channels have the same mesh. Regards, JP -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Wed Nov 25 11:05:06 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 25 Nov 2015 18:05:06 +0100 Subject: [Nek5000-users] LES of a jet - Vreman model Message-ID: Dear Neks, I try to implement the Vreman LES model (see An eddy-viscosity subgrid-scale model for turbulent shear flow: Algebraic theory and applications, A. W. Vreman, Phys. Fl., 16, 10, 2004) to simulate a turbulent jet. To do so, I modified the turbChannel example provided in NEK. However, as soon as I impose in the uservp subroutine a space-varying viscosity, after some time steps, the simulation blows up even if this viscosity remains close from the standard 1/Re viscosity. Has anyone ever experienced this kind of issue? Does anyone have an idea on how to fix this problem? Below are the .usr file and the log of the simulation. Thanks, Best regards, L?opold ------------------------------------------- .usr file c----------------------------------------------------------------------- c c Based on the turbChannel example c c----------------------------------------------------------------------- subroutine uservp (ix,iy,iz,eg) include 'SIZE' include 'TOTAL' include 'NEKUSE' common /cdsmag/ ediff(lx1,ly1,lz1,lelv) integer ie,f,eg real aux ie = gllel(eg) udiff = ediff(ix,iy,iz,ie) c udiff = param(2) utrans=1. aux = ediff(ix,iy,iz,ie) - param(2) if (istep.eq.19) then write(6,*) aux endif return end c----------------------------------------------------------------------- subroutine userf (ix,iy,iz,eg) include 'SIZE' include 'TOTAL' include 'NEKUSE' integer e,f,eg c e = gllel(eg) ffx = 0.0 ffy = 0.0 ffz = 0.0 return end c----------------------------------------------------------------------- subroutine userq (ix,iy,iz,ieg) include 'SIZE' include 'TOTAL' include 'NEKUSE' qvol = 0. return end c----------------------------------------------------------------------- subroutine userchk include 'SIZE' include 'TOTAL' include 'RESTART' include 'ZPER' ! for nelx,nely,nelz parameter(lt=lx1*ly1*lz1*lelv) real vort(lt,3),vort_pol(lt,3),w1(lt),w2(lt),rayon real x0(3) save x0 integer icalld save icalld data icalld /0/ character*1 snam1(80) character*1 f1nam1(80),f2nam1(80) character*80 f1name equivalence (f1nam1,f1name) character*80 f2name equivalence (f2nam1,f2name) real atime,timel save atime,timel integer icount save icount common /cdsmag/ ediff(lx1,ly1,lz1,lelv) common /avg/ uravg(lx1,ly1,lz1,lelv) & , utavg(lx1,ly1,lz1,lelv) & , uzavg(lx1,ly1,lz1,lelv) common /lesleo/ sij (lx1*ly1*lz1,ldim,ldim) $ , dx2 (lx1,ly1,lz1,lelv,lp) $ , dy2 (lx1,ly1,lz1,lelv,lp) $ , dz2 (lx1,ly1,lz1,lelv,lp) $ , cdiff real sij,dx2,dy2,dz2,cdiff integer e logical ifverbose n=nx1*ny1*nz1*nelv pi = 4.*atan(1.0) rho = 1. dnu = param(2) cdiff = 0.07 ! add SGS term to RHS (explicit treatment in plan4) ! Explicit viscosity. if(ifsplit .and. param(30).gt.0) ifexplvis = .true. c Grid spacing. c if ((istep.lt.1).and.(nid.eq.0)) then c if (istep.lt.1) then if (nid.eq.(istep-1)) then call set_grid_spacing(dx2,dy2,dz2) endif ! Compute eddy viscosity using Vreman model c Waiting for several timesteps in order to be sure that gradients c are not zero if (istep.gt.17) then ifuservp = .true. else ifuservp = .false. endif if(istep.gt.16) then if(nid.eq.0) write(6,*) 'Calculating eddy visosity' do e=1,nelv call eddy_visc(ediff,e) enddo c call copy(t,ediff,n) endif c c Below is just for postprocessing ... c c Compute perturbation energy e2 = glsc3(vy,bm1,vy,n)+glsc3(vz,bm1,vz,n) e2 = e2/volvm1 ifxyo = .true. ! Turn on xyz output if (istep.gt.iostep) ifxyo = .false. ! Turn off xyz output after first dump c Average calculations. if(icalld.eq.0) then call rzero(uravg,n) call rzero(utavg,n) call rzero(uzavg,n) atime = 0. timel = time icalld = 1 endif dtime = time - timel atime = atime + dtime if (atime.ne.0. .and. dtime.ne.0.) then beta = dtime/atime alpha = 1.-beta ifverbose = .false. call avg1(uravg,vx ,alpha,beta,n,'uravg',ifverbose) call avg1(utavg,vy ,alpha,beta,n,'utavg',ifverbose) call avg1(uzavg,vz ,alpha,beta,n,'uzavg',ifverbose) endif timel = time ntot = nx1*ny1*nz1*nelv c Vorticity calculations. if (mod(istep,iostep).eq.0) then call comp_vort3(vort,w1,w2,vx,vy,vz) c Calculations done in polar coordinates. do i=1,ntot rayon = SQRT(xm1(i,1,1,1)**2 + ym1(i,1,1,1)**2) if (rayon>0.0) then vort_pol(i,1)=(xm1(i,1,1,1)*vort(i,1) + $ ym1(i,1,1,1)*vort(i,2))/rayon vort_pol(i,2)=(-ym1(i,1,1,1)*vort(i,1) + $ xm1(i,1,1,1)*vort(i,2))/rayon vort_pol(i,3)=vort(i,3) else vort_pol(i,1)=0.0 vort_pol(i,2)=0.0 vort_pol(i,3)=vort(i,3) endif enddo c call outpost(vort(1,1),vort(1,2),vort(1,3), c $ pr,t(1,1,1,1,1),'vrt') call outpost(vort_pol(1,1),vort_pol(1,2),vort_pol(1,3), $ pr,t(1,1,1,1,1),'pol') endif if (istep.eq.1) then endif c We export the grid spacing and the average values of velocities. if(istep.eq.18) then ifxyo = .true. call outpost(dx2(1,1,1,1,1),dy2(1,1,1,1,1),dz2(1,1,1,1,1), $ ediff(1,1,1,1),t(1,1,1,1,1),'les') call outpost(uravg(1,1,1,1),utavg(1,1,1,1),uzavg(1,1,1,1), $ pr,t(1,1,1,1,1),'avg') if (nid.eq.0) then open(223,file='diffusion.dat',status='old',action='write', $ form='formatted',position="append") do i=1,ntot c write(223,*) xm1(i,1,1,1),ym1(i,1,1,1),zm1(i,1,1,1),nid, c $ ediff(i,1,1,1),dx2(i,1,1,1,nid+1),dy2(i,1,1,1,nid+1), c $ dz2(i,1,1,1,nid+1) write(223,*) nid enddo close(223) endif if (nid.eq.1) then open(225,file='diffusion1.dat',status='old',action='write', $ form='formatted',position="append") do i=1,ntot c write(223,*) xm1(i,1,1,1),ym1(i,1,1,1),zm1(i,1,1,1),nid, c $ ediff(i,1,1,1),dx2(i,1,1,1,nid+1),dy2(i,1,1,1,nid+1), c $ dz2(i,1,1,1,nid+1) write(225,*) nid enddo close(225) endif endif return end c----------------------------------------------------------------------- subroutine userbc (ix,iy,iz,iside,ieg) c NOTE ::: This subroutine MAY NOT be called by every process include 'SIZE' include 'TOTAL' include 'NEKUSE' c ux=0.0 uy=0.0 uz=0.0 temp=0 c z0 = 0.05 rayon = SQRT(x**2 + y**2) ! Radius in polar coordinates R = 1.0 ! Radius of the jet theta0 = 0.05*R ! Thickness of the mixing layer axialamp = 0.05 ! Turbulence amplitude (axial excitation) helicamp = 0.05 ! Turbulence amplitude (helical direction) faxial = 0.55 ! Axial frequency = St*U0/D et D=1.0 fhelic = faxial/2.0 ! Helical frequency = faxial/2.0 phihelic = 0.0 ! Phase lag between axial and helical c if (z.le.z0) then ux=0.0 uy=0.0 if (rayon.gt.(0.0)) then uz=0.5*(1-tanh(0.25*(rayon/R - R/rayon)*R/theta0)) else uz=1.0 endif endif c return end c----------------------------------------------------------------------- subroutine useric (ix,iy,iz,ieg) include 'SIZE' include 'TOTAL' include 'NEKUSE' rayon = SQRT(x**2 + y**2) ! Radius in polar coordinates R = 0.5 ! Radius of the jet theta0 = 0.05*R ! Thickness of the mixing layer ux=0.0 uy=0.0 if (rayon.gt.(0.0)) then uz=0.5*(1-tanh(0.25*(rayon/R - R/rayon)*R/theta0)) else uz=1.0 endif temp=0 return end c----------------------------------------------------------------------- subroutine usrdat include 'SIZE' include 'TOTAL' common /cdsmag/ ediff(lx1,ly1,lz1,lelv) n = nx1*ny1*nz1*nelt call cfill(ediff,param(2),n) ! initialize viscosity ! enable stress formulation if we use PN-PN-2 if(lx1.ne.lx2 .and. param(30).gt.0) IFSTRS = .true. return end c----------------------------------------------------------------------- subroutine usrdat2 include 'SIZE' include 'TOTAL' return end c----------------------------------------------------------------------- subroutine usrdat3 include 'SIZE' include 'TOTAL' c return end c----------------------------------------------------------------------- subroutine set_obj ! define objects for surface integrals c include 'SIZE' include 'TOTAL' integer e,f,eg nobj = 1 iobj = 0 do ii=nhis+1,nhis+nobj iobj = iobj+1 hcode(10,ii) = 'I' hcode( 1,ii) = 'F' hcode( 2,ii) = 'F' hcode( 3,ii) = 'F' lochis(1,ii) = iobj enddo nhis = nhis + nobj if (maxobj.lt.nobj) call exitti('increase maxobj in SIZE$',nobj) nxyz = nx1*ny1*nz1 nface = 2*ndim do e=1,nelv do f=1,nface if (cbc(f,e,1).eq.'W ') then iobj = 1 if (iobj.gt.0) then nmember(iobj) = nmember(iobj) + 1 mem = nmember(iobj) eg = lglel(e) object(iobj,mem,1) = eg object(iobj,mem,2) = f c write(6,1) iobj,mem,f,eg,e,nid,' OBJ' c 1 format(6i9,a4) endif endif enddo enddo c write(6,*) 'number',(nmember(k),k=1,4) c return end c----------------------------------------------------------------------- subroutine eddy_visc(ediff,e) c c Compute eddy viscosity using Vreman's model. c include 'SIZE' include 'TOTAL' include 'ZPER' real ediff(nx1*ny1*nz1,nelv) real betaij(lx1*ly1*lz1,ldim,ldim) real Bbeta(lx1*ly1*lz1) real Sommeij2(lx1*ly1*lz1) integer e,i,ldim1,ldim2 common /lesleo/ sij (lx1*ly1*lz1,ldim,ldim) $ , dx2 (lx1*ly1*lz1,lelv,lp) $ , dy2 (lx1*ly1*lz1,lelv,lp) $ , dz2 (lx1*ly1*lz1,lelv,lp) $ , cdiff real sij,dx2,dy2,dz2,cdiff integer ntot ntot = nx1*ny1*nz1 c Velocity gradient. call comp_gije(sij,vx(1,1,1,e),vy(1,1,1,e),vz(1,1,1,e),e) c Calculation of Beta. do ldim2=1,ldim do ldim1=1,ldim do i=1,ntot betaij(i,ldim1,ldim2) = $ dx2(i,e,nid+1)*sij(i,1,ldim1)*sij(i,1,ldim2) $ + dy2(i,e,nid+1)*sij(i,2,ldim1)*sij(i,2,ldim2) $ + dz2(i,e,nid+1)*sij(i,3,ldim1)*sij(i,3,ldim2) enddo enddo enddo c Calculation of Bbeta. do i=1,ntot Bbeta(i) = betaij(i,1,1)*betaij(i,2,2) - (betaij(i,1,2)**2) $ + betaij(i,2,2)*betaij(i,3,3) - (betaij(i,2,3)**2) $ + betaij(i,3,3)*betaij(i,1,1) - (betaij(i,3,1)**2) enddo c Calculons Sommeij2 = Sum_{i,j} (alpha_{ij}^2) pour le mod?le de c Vreman. On met d'abord Sommeij2 ? zero. call rzero(Sommeij2,ntot) do ldim2=1,ldim do ldim1=1,ldim do i=1,ntot Sommeij2(i) = Sommeij2(i) + (sij(i,ldim1,ldim2)**2) enddo enddo enddo c ediff calculation. do i=1,ntot if (Sommeij2(i).gt.0.0) then ediff(i,e) = param(2) + cdiff*sqrt(Bbeta(i)/Sommeij2(i)) c ediff(i,e) = param(2) else ediff(i,e) = param(2) endif enddo return end c----------------------------------------------------------------------- subroutine set_grid_spacing(dx2,dy2,dz2) c c Compute D^2, the grid spacing used in the DS sgs model. c include 'SIZE' include 'TOTAL' real dx2(nx1,ny1,nz1,nelv,np) real dy2(nx1,ny1,nz1,nelv,np) real dz2(nx1,ny1,nz1,nelv,np) real longueur,largeur,dhlong,dhlarg,deltai,deltaj integer e,i,j,k,im,jm,km,ip,jp,kp c return ! Comment this line for a non-trivial Delta defn write(6,*) 'LadHyX' open (224,file='dx2dy2dz2.dat',status='old',action='write', $ form='formatted',position="append") do e=1,nelv do k=1,nz1 km = max(1 ,k-1) kp = min(nz1,k+1) do j=1,ny1 jm = max(1 ,j-1) jp = min(ny1,j+1) do i=1,nx1 im = max(1 ,i-1) ip = min(nx1,i+1) longueur = 0.5*(abs(xm1(ip,jp,k,e) - xm1(im,jp,k,e)) $ + abs(xm1(ip,jm,k,e) - xm1(im,jm,k,e)))/(ip - im) largeur = 0.5*(abs(ym1(ip,jp,k,e) - ym1(ip,jm,k,e)) $ + abs(ym1(im,jp,k,e) - ym1(im,jm,k,e)))/(jp - jm) dhlong = 0.5*(abs(ym1(ip,jp,k,e) - ym1(im,jp,k,e)) $ + abs(ym1(ip,jm,k,e) - ym1(im,jm,k,e)))/(ip - im) dhlarg = 0.5*(abs(xm1(ip,jp,k,e) - xm1(ip,jm,k,e)) $ + abs(xm1(im,jp,k,e) - xm1(im,jm,k,e)))/(jp - jm) deltai = sqrt(longueur**2 + dhlong**2) deltaj = sqrt(largeur**2 + dhlarg**2) dx2(i,j,k,e,nid+1) = ((longueur**2)/deltai + (dhlarg**2)/ $ deltaj)**2 dy2(i,j,k,e,nid+1) = ((dhlong**2)/deltai + (largeur**2)/ $ deltaj)**2 dz2(i,j,k,e,nid+1) = ((zm1(i,j,kp,e)-zm1(i,j,km,e))/ $ (kp-km))**2 write(224,*) i,j,k,e,nid,dx2(i,j,k,e,nid+1), $ dy2(i,j,k,e,nid+1),dz2(i,j,k,e,nid+1) enddo enddo enddo enddo close(224) return end c----------------------------------------------------------------------- c c automatically added by makenek subroutine usrsetvert(glo_num,nel,nx,ny,nz) ! to modify glo_num integer*8 glo_num(1) return end ------------------------------------------- log file /----------------------------------------------------------\\ | _ __ ______ __ __ ______ ____ ____ ____ | | / | / // ____// //_/ / ____/ / __ \\ / __ \\ / __ \\ | | / |/ // __/ / ,< /___ \\ / / / // / / // / / / | | / /| // /___ / /| | ____/ / / /_/ // /_/ // /_/ / | | /_/ |_//_____//_/ |_|/_____/ \\____/ \\____/ \\____/ | | | |----------------------------------------------------------| | | | NEK5000: Open Source Spectral Element Solver | | COPYRIGHT (c) 2008-2010 UCHICAGO ARGONNE, LLC | | Version: 1.0rc1 / SVN r1088 | | Web: http://nek5000.mcs.anl.gov | | | \\----------------------------------------------------------/ Number of processors: 16 REAL wdsize : 8 INTEGER wdsize : 4 Beginning session: /home/leopold/NEK5000/turbessai_2511_1714/jet3d.rea timer accuracy: 0.0000000E+00 sec read .rea file nelgt/nelgv/lelt: 1200 1200 75 lx1 /lx2 /lx3 : 8 6 6 mapping elements to processors 1 75 75 1200 1200 NELV 2 75 75 1200 1200 NELV 3 75 75 1200 1200 NELV 4 75 75 1200 1200 NELV 5 75 75 1200 1200 NELV 6 75 75 1200 1200 NELV 7 75 75 1200 1200 NELV 8 75 75 1200 1200 NELV 9 75 75 1200 1200 NELV 10 75 75 1200 1200 NELV 11 75 75 1200 1200 NELV 12 75 75 1200 1200 NELV 13 75 75 1200 1200 NELV 14 75 75 1200 1200 NELV 15 75 75 1200 1200 NELV 0 75 75 1200 1200 NELV RANK 0 IEG 851 861 871 951 952 953 954 961 962 963 964 965 971 972 973 974 975 981 982 983 984 985 991 992 993 994 995 1051 1052 1053 1054 1061 1062 1063 1064 1065 1071 1072 1073 1074 1075 1081 1082 1083 1084 1085 1091 1092 1093 1094 1095 1151 1152 1153 1154 1161 1162 1163 1164 1165 1171 1172 1173 1174 1175 1181 1182 1183 1184 1185 1191 1192 1193 1194 1195 element load imbalance: 0 75 75 done :: mapping elements to processors 0 objects found done :: read .rea file 0.32928E-01 sec setup mesh topology Right-handed check complete for 1200 elements. OK. setvert3d: 8 169285 428485 169285 169285 call usrsetvert done :: usrsetvert gs_setup: 30185 unique labels shared pairwise times (avg, min, max): 8.32036e-05 8.0204e-05 8.54015e-05 crystal router : 0.00022305 0.000220394 0.000225186 all reduce : 0.000695395 0.000693202 0.000698304 used all_to_all method: pairwise handle bytes (avg, min, max): 297659 277636 321556 buffer bytes (avg, min, max): 65870 50368 84512 setupds time 1.1183E-01 seconds 0 8 169285 1200 8 max multiplicity done :: setup mesh topology call usrdat done :: usrdat generate geometry data vol_t,vol_v: 76.800000000007586 76.800000000007586 done :: generate geometry data call usrdat2 done :: usrdat2 regenerate geometry data 1 vol_t,vol_v: 76.800000000007373 76.800000000007373 done :: regenerate geometry data 1 verify mesh topology -2.0000000000000018 2.0000000000000018 Xrange -2.0000000000000018 2.0000000000000018 Yrange 0.0000000000000000 4.8000000000000069 Zrange done :: verify mesh topology 103 Parameters from file:/home/leopold/NEK5000/turbessai_2511_1714/jet3d.rea 1 1.000000 p01 DENSITY 2 -600.000 p02 VISCOS 7 1.000000 p07 RHOCP 8 -1492.000 p08 CONDUCT 11 300.00 p11 NSTEPS 12 4.0000000E-02 p12 DT 15 25.000 p15 IOSTEP 21 1.0000000E-07 p21 DIVERGENCE 22 8.0000000E-09 p22 HELMHOLTZ 24 5.0000000E-03 p24 TOLREL 25 1.0000000E-02 p25 TOLABS 26 2.000000 p26 COURANT/NTAU 27 2.000000 p27 TORDER 28 0.0000000E+00 p28 TORDER: mesh velocity (0: p28=p27) 30 1.0000000E+00 p30 > 0 ==> properties set in uservp() 65 1.000000 p65 #iofile(eg,0 or 64); <0 --> sep. dirs 66 6.000000 p66 write fmt:ONLY postx uses rea value 67 6.000000 p67 read fmt: same modes as p66 68 1000.000 p68 iastep: freq for avg_all 93 30.00000 p93 Numbr of prev pressure solns saved 99 3.000000 p99 dealiasing:if <0 disable 103 1.0000000E-03 p103 weight of stabilizing filter (.01) IFTRAN = T IFFLOW = T IFHEAT = F IFSPLIT = F IFLOMACH = F IFUSERVP = T IFUSERMV = F IFSTRS = T IFCHAR = T IFCYCLIC = F IFAXIS = F IFMVBD = F IFMELT = F IFMODEL = F IFKEPS = F IFMOAB = F IFNEKNEK = F IFSYNC = T IFVCOR = F IFINTQ = F IFCWUZ = F IFSWALL = F IFGEOM = F IFSURT = F IFWCNO = F IFCMT = F IFVISC = F IFFLTR = F IFTMSH for field 1 = F IFADVC for field 1 = T IFNONL for field 1 = F Dealiasing enabled, lxd= 12 Estimated eigenvalues EIGAA = 0.55402293223707511 EIGGA = 43231.464629758724 EIGAE = 0.42836824657505768 EIGAS = 6.4432989690721601E-002 EIGGE = 43231.464629758724 EIGGS = 2.0000000000000000 verify mesh topology -2.0000000000000018 2.0000000000000018 Xrange -2.0000000000000018 2.0000000000000018 Yrange 0.0000000000000000 4.8000000000000069 Zrange done :: verify mesh topology E-solver strategy: 1 itr mg_nx: 1 3 7 mg_ny: 1 3 7 mg_nz: 1 3 7 call usrsetvert done :: usrsetvert gs_setup: 641 unique labels shared pairwise times (avg, min, max): 5.16847e-05 5.04971e-05 5.30005e-05 crystal router : 2.23234e-05 2.19107e-05 2.26021e-05 all reduce : 4.48287e-05 4.42982e-05 4.52042e-05 used all_to_all method: crystal router handle bytes (avg, min, max): 22825 20364 25796 buffer bytes (avg, min, max): 3360 3040 3744 setupds time 3.4559E-03 seconds 1 2 1573 1200 setvert3d: 4 25957 35557 25957 25957 call usrsetvert done :: usrsetvert gs_setup: 5593 unique labels shared pairwise times (avg, min, max): 4.74274e-05 4.41074e-05 4.97103e-05 crystal router : 5.89028e-05 5.86033e-05 5.95093e-05 all reduce : 0.000171191 0.000170612 0.000171995 used all_to_all method: pairwise handle bytes (avg, min, max): 58603 54564 63444 buffer bytes (avg, min, max): 13614 10304 17696 setupds time 9.9120E-03 seconds 2 4 25957 1200 setvert3d: 4 25957 35557 25957 25957 call usrsetvert done :: usrsetvert gs_setup: 5593 unique labels shared pairwise times (avg, min, max): 4.033e-05 3.91006e-05 4.20094e-05 crystal router : 5.31495e-05 5.26905e-05 5.39064e-05 all reduce : 0.000171348 0.000170612 0.000172019 used all_to_all method: pairwise handle bytes (avg, min, max): 58603 54564 63444 buffer bytes (avg, min, max): 13614 10304 17696 setupds time 9.2380E-03 seconds 3 4 25957 1200 setvert3d: 6 81861 158661 81861 81861 call usrsetvert done :: usrsetvert gs_setup: 15441 unique labels shared pairwise times (avg, min, max): 6.1132e-05 6.02007e-05 6.27041e-05 crystal router : 0.00013117 0.00013051 0.000131893 all reduce : 0.000425158 0.000423312 0.000426292 used all_to_all method: pairwise handle bytes (avg, min, max): 154899 144404 167444 buffer bytes (avg, min, max): 34846 26560 44896 setupds time 2.1295E-02 seconds 4 6 81861 1200 setup h1 coarse grid, nx_crs= 2 call usrsetvert done :: usrsetvert gs_setup: 641 unique labels shared pairwise times (avg, min, max): 5.51209e-05 5.09977e-05 5.77927e-05 crystal router : 1.62676e-05 1.59979e-05 1.65939e-05 all reduce : 5.0886e-05 4.99964e-05 5.14984e-05 used all_to_all method: crystal router handle bytes (avg, min, max): 22825 20364 25796 buffer bytes (avg, min, max): 3360 3040 3744 done :: setup h1 coarse grid 2.9180049896240234E-002 sec call usrdat3 done :: usrdat3 set initial conditions nekuic (1) for ifld 1 call nekuic for vel xyz min -2.0000 -2.0000 0.0000 uvwpt min 0.0000 0.0000 0.0000 0.0000 0.0000 PS min 0.0000 0.0000 0.99000E+22 xyz max 2.0000 2.0000 4.8000 uvwpt max 0.0000 0.0000 1.0000 0.0000 0.0000 PS max 0.0000 0.0000 -0.99000E+22 done :: set initial conditions call userchk schfile:/home/leopold/NEK5000/turbessai_2511_1714/jet3d.sch call outfld: ifpsco: F 0 0.0000E+00 Write checkpoint: 0 0 OPEN: poljet3d0.f00001 0 0.0000E+00 done :: Write checkpoint file size = 17. MB avg data-throughput = 520.9MB/s io-nodes = 1 done :: userchk gridpoints unique/tot: 428485 614400 dofs: 423444 259200 Initialization successfully completed 0.99081 sec Starting time loop ... DT/DTCFL/DTFS/DTINIT 0.400E-01 0.513E-01 0.000E+00 0.400E-01 Step 1, t= 4.0000000E-02, DT= 4.0000000E-02, C= 1.559 0.0000E+00 0.0000E+00 Solving for fluid F T T 0 8.0000E-09 1.0338E-06 8.0000E-09 tol,matmod 1 Helmholtz3/fluid: 11 0.5267E-08 0.8000E-08 0.1797E-01 1 1.00000E-07 4.40911E-01 9.30604E-01 4.73790E-01 1 Divergence 2 1.00000E-07 1.76488E-01 9.30604E-01 1.89649E-01 1 Divergence 3 1.00000E-07 5.55944E-02 9.30604E-01 5.97401E-02 1 Divergence 4 1.00000E-07 1.75339E-02 9.30604E-01 1.88415E-02 1 Divergence 5 1.00000E-07 6.73386E-03 9.30604E-01 7.23600E-03 1 Divergence 6 1.00000E-07 2.27044E-03 9.30604E-01 2.43975E-03 1 Divergence 7 1.00000E-07 7.55678E-04 9.30604E-01 8.12029E-04 1 Divergence 8 1.00000E-07 2.71533E-04 9.30604E-01 2.91781E-04 1 Divergence 9 1.00000E-07 9.26116E-05 9.30604E-01 9.95177E-05 1 Divergence 10 1.00000E-07 2.89531E-05 9.30604E-01 3.11121E-05 1 Divergence 11 1.00000E-07 1.00088E-05 9.30604E-01 1.07552E-05 1 Divergence 12 1.00000E-07 4.02671E-06 9.30604E-01 4.32698E-06 1 Divergence 13 1.00000E-07 1.70162E-06 9.30604E-01 1.82852E-06 1 Divergence 14 1.00000E-07 6.61159E-07 9.30604E-01 7.10462E-07 1 Divergence 15 1.00000E-07 2.33259E-07 9.30604E-01 2.50654E-07 1 Divergence 16 1.00000E-07 7.88484E-08 9.30604E-01 8.47282E-08 1 Divergence 1 U-PRES gmres: 16 7.8848E-08 1.0000E-07 9.3060E-01 7.7317E-02 3.0742E-01 1 4.0000E-02 5.1049E-01 Fluid done filt amp 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0010 filt trn 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 0.9990 LadHyX DT/DTCFL/DTFS/DTINIT 0.329E-01 0.329E-01 0.000E+00 0.400E-01 Step 2, t= 7.2887311E-02, DT= 3.2887311E-02, C= 1.676 1.1179E+00 1.1179E+00 Solving for fluid F T T 2 Helmholtz3/fluid: 11 0.2403E-08 0.8000E-08 0.3008E+00 2 U-PRES gmres: 16 6.7074E-08 1.0000E-07 6.6007E-01 7.7229E-02 2.9921E-01 2 7.2887E-02 5.0106E-01 Fluid done LadHyX Step 3, t= 1.0577462E-01, DT= 3.2887311E-02, C= 1.686 2.5436E+00 1.4257E+00 Solving for fluid F T T 3 Helmholtz3/fluid: 10 0.3690E-08 0.8000E-08 0.9894E-01 3 U-PRES gmres: 15 6.4098E-08 1.0000E-07 2.0530E-01 6.6209E-02 2.5669E-01 3 1.0577E-01 4.3805E-01 Fluid done LadHyX Step 4, t= 1.3866193E-01, DT= 3.2887311E-02, C= 1.693 3.8424E+00 1.2988E+00 Solving for fluid F T T 4 Helmholtz3/fluid: 9 0.7802E-08 0.8000E-08 0.3080E-01 4 U-PRES gmres: 12 8.7915E-08 1.0000E-07 4.0701E-02 5.3055E-02 2.0200E-01 4 1.3866E-01 3.6896E-01 Fluid done LadHyX Step 5, t= 1.7154924E-01, DT= 3.2887311E-02, C= 1.699 5.0340E+00 1.1916E+00 Solving for fluid F T T 5 Helmholtz3/fluid: 9 0.7287E-08 0.8000E-08 0.2859E-01 5 U-PRES gmres: 12 6.0219E-08 1.0000E-07 2.5639E-02 5.3179E-02 2.0134E-01 5 1.7155E-01 3.6834E-01 Fluid done LadHyX Step 6, t= 2.0443656E-01, DT= 3.2887311E-02, C= 1.695 6.2218E+00 1.1877E+00 Solving for fluid F T T 6 Helmholtz3/fluid: 9 0.6680E-08 0.8000E-08 0.2745E-01 6 U-PRES gmres: 12 6.6203E-08 1.0000E-07 3.2161E-02 5.2715E-02 2.0061E-01 6 2.0444E-01 3.6690E-01 Fluid done LadHyX Step 7, t= 2.3732387E-01, DT= 3.2887311E-02, C= 1.697 7.4158E+00 1.1941E+00 Solving for fluid F T T 7 Helmholtz3/fluid: 9 0.6353E-08 0.8000E-08 0.2680E-01 7 U-PRES gmres: 12 5.1712E-08 1.0000E-07 2.6367E-02 5.2906E-02 2.0242E-01 7 2.3732E-01 3.6932E-01 Fluid done LadHyX Step 8, t= 2.7021118E-01, DT= 3.2887311E-02, C= 1.698 8.6288E+00 1.2129E+00 Solving for fluid F T T 8 Helmholtz3/fluid: 9 0.6673E-08 0.8000E-08 0.2685E-01 8 U-PRES gmres: 12 4.4087E-08 1.0000E-07 2.3825E-02 5.2858E-02 2.0136E-01 8 2.7021E-01 3.6842E-01 Fluid done LadHyX Step 9, t= 3.0309849E-01, DT= 3.2887311E-02, C= 1.684 9.8241E+00 1.1954E+00 Solving for fluid F T T 9 Helmholtz3/fluid: 9 0.7127E-08 0.8000E-08 0.2677E-01 9 U-PRES gmres: 12 4.5466E-08 1.0000E-07 2.2156E-02 5.2914E-02 2.0164E-01 9 3.0310E-01 3.6852E-01 Fluid done LadHyX Step 10, t= 3.3598580E-01, DT= 3.2887311E-02, C= 1.645 1.1023E+01 1.1992E+00 Solving for fluid F T T 10 Helmholtz3/fluid: 9 0.6816E-08 0.8000E-08 0.2582E-01 10 U-PRES gmres: 12 4.4157E-08 1.0000E-07 1.8823E-02 5.2885E-02 2.0112E-01 10 3.3599E-01 3.6791E-01 Fluid done LadHyX Step 11, t= 3.6887311E-01, DT= 3.2887311E-02, C= 1.632 1.2209E+01 1.1855E+00 Solving for fluid F T T 11 Helmholtz3/fluid: 9 0.6294E-08 0.8000E-08 0.2484E-01 11 U-PRES gmres: 12 4.1769E-08 1.0000E-07 1.6670E-02 5.2812E-02 2.0193E-01 11 3.6887E-01 3.6934E-01 Fluid done LadHyX Step 12, t= 4.0176042E-01, DT= 3.2887311E-02, C= 1.641 1.3405E+01 1.1956E+00 Solving for fluid F T T 12 Helmholtz3/fluid: 9 0.6230E-08 0.8000E-08 0.2475E-01 12 U-PRES gmres: 12 4.2914E-08 1.0000E-07 1.6074E-02 5.2975E-02 2.0160E-01 12 4.0176E-01 3.6909E-01 Fluid done LadHyX Step 13, t= 4.3464773E-01, DT= 3.2887311E-02, C= 1.651 1.4600E+01 1.1958E+00 Solving for fluid F T T 13 Helmholtz3/fluid: 9 0.6215E-08 0.8000E-08 0.2488E-01 13 U-PRES gmres: 12 3.9520E-08 1.0000E-07 1.4803E-02 5.2862E-02 2.0123E-01 13 4.3465E-01 3.6785E-01 Fluid done LadHyX Step 14, t= 4.6753504E-01, DT= 3.2887311E-02, C= 1.661 1.5797E+01 1.1967E+00 Solving for fluid F T T 14 Helmholtz3/fluid: 9 0.6096E-08 0.8000E-08 0.2437E-01 14 U-PRES gmres: 12 3.6261E-08 1.0000E-07 1.3651E-02 5.3142E-02 2.0131E-01 14 4.6754E-01 3.6796E-01 Fluid done LadHyX Step 15, t= 5.0042236E-01, DT= 3.2887311E-02, C= 1.671 1.6990E+01 1.1930E+00 Solving for fluid F T T 15 Helmholtz3/fluid: 9 0.5814E-08 0.8000E-08 0.2304E-01 15 U-PRES gmres: 12 3.8622E-08 1.0000E-07 1.2689E-02 5.2549E-02 2.0080E-01 15 5.0042E-01 3.6748E-01 Fluid done LadHyX Step 16, t= 5.3330967E-01, DT= 3.2887311E-02, C= 1.686 1.8185E+01 1.1949E+00 Solving for fluid F T T 16 Helmholtz3/fluid: 9 0.5392E-08 0.8000E-08 0.2159E-01 16 U-PRES gmres: 12 4.0755E-08 1.0000E-07 1.1259E-02 5.2697E-02 2.0102E-01 16 5.3331E-01 3.6759E-01 Fluid done LadHyX Step 17, t= 5.6619698E-01, DT= 3.2887311E-02, C= 1.706 1.9387E+01 1.2019E+00 Solving for fluid F T T 17 Helmholtz3/fluid: 9 0.5216E-08 0.8000E-08 0.2097E-01 17 U-PRES gmres: 12 3.9763E-08 1.0000E-07 1.0706E-02 5.2939E-02 2.0170E-01 17 5.6620E-01 3.6837E-01 Fluid done Calculating eddy visosity Step 18, t= 5.9908429E-01, DT= 3.2887311E-02, C= 1.725 2.0349E+01 9.6170E-01 Solving for fluid F T T 18 Helmholtz3/fluid: 9 0.5243E-08 0.8000E-08 0.2106E-01 18 U-PRES gmres: 12 3.4908E-08 1.0000E-07 1.0308E-02 5.3016E-02 2.0171E-01 18 5.9908E-01 3.6860E-01 Fluid done Calculating eddy visosity call outfld: ifpsco: F 18 5.9908E-01 Write checkpoint: 0 18 OPEN: lesjet3d0.f00001 18 5.9908E-01 done :: Write checkpoint file size = 17. MB avg data-throughput = 617.1MB/s io-nodes = 1 call outfld: ifpsco: F 18 5.9908E-01 Write checkpoint: 0 18 OPEN: avgjet3d0.f00001 18 5.9908E-01 done :: Write checkpoint file size = 17. MB avg data-throughput = 647.7MB/s io-nodes = 1 Step 19, t= 6.3197160E-01, DT= 3.2887311E-02, C= 1.739 2.1382E+01 1.0337E+00 Solving for fluid F T T 19 Helmholtz3/fluid: 10 0.1805E-08 0.8000E-08 0.2278E-01 19 U-PRES gmres: 11 9.1061E-08 1.0000E-07 9.2230E-03 5.2888E-02 2.0412E-01 19 6.3197E-01 4.0092E-01 Fluid done Calculating eddy visosity Step 20, t= 6.6485891E-01, DT= 3.2887311E-02, C= 1.732 2.5370E+01 3.9873E+00 Solving for fluid F T T 20 Helmholtz3/fluid: 10 0.1673E-08 0.8000E-08 0.2231E-01 20 U-PRES gmres: 11 8.8236E-08 1.0000E-07 8.5941E-03 5.2522E-02 2.0177E-01 20 6.6486E-01 3.8955E-01 Fluid done Calculating eddy visosity Step 21, t= 6.9774622E-01, DT= 3.2887311E-02, C= 1.700 2.6416E+01 1.0463E+00 Solving for fluid F T T 21 Helmholtz3/fluid: 10 0.1618E-08 0.8000E-08 0.2169E-01 21 U-PRES gmres: 11 8.8444E-08 1.0000E-07 8.0928E-03 5.2019E-02 1.9286E-01 21 6.9775E-01 3.8227E-01 Fluid done Calculating eddy visosity ...................................................... Step 245, t= 8.6498981E+00, DT= 3.9464773E-02, C= 1.878 2.4108E+02 9.7130E-01 Solving for fluid F T T 245 Helmholtz3/fluid: 11 0.1876E-08 0.8000E-08 0.3329E-01 245 U-PRES gmres: 12 4.0605E-08 1.0000E-07 1.2480E-02 5.3032E-02 2.0306E-01 245 8.6499E+00 3.9528E-01 Fluid done Calculating eddy visosity Step 246, t= 8.6893628E+00, DT= 3.9464773E-02, C= 1.889 2.4207E+02 9.8505E-01 Solving for fluid F T T 246 Helmholtz3/fluid: 11 0.1944E-08 0.8000E-08 0.3391E-01 246 U-PRES gmres: 12 4.5293E-08 1.0000E-07 1.4596E-02 5.3483E-02 2.0330E-01 246 8.6894E+00 3.9605E-01 Fluid done Calculating eddy visosity Step 247, t= 8.7288276E+00, DT= 3.9464773E-02, C= 1.898 2.4305E+02 9.8502E-01 Solving for fluid F T T 247 Helmholtz3/fluid: 11 0.2644E-08 0.8000E-08 0.3535E-01 247 U-PRES gmres: 12 8.9201E-08 1.0000E-07 2.1450E-02 5.7194E-02 2.1392E-01 247 8.7288E+00 4.1121E-01 Fluid done Calculating eddy visosity Step 248, t= 8.7682924E+00, DT= 3.9464773E-02, C= 1.906 2.4411E+02 1.0537E+00 Solving for fluid F T T 248 Helmholtz3/fluid: 11 0.4365E-08 0.8000E-08 0.4022E-01 248 U-PRES gmres: 12 8.4866E-08 1.0000E-07 2.8833E-02 5.6473E-02 2.1303E-01 248 8.7683E+00 4.1053E-01 Fluid done Calculating eddy visosity Step 249, t= 8.8077571E+00, DT= 3.9464773E-02, C= 2.072 2.4516E+02 1.0495E+00 Solving for fluid F T T 249 Helmholtz3/fluid: 12 0.1785E-08 0.8000E-08 0.6760E-01 249 U-PRES gmres: 13 8.3312E-08 1.0000E-07 1.0799E-01 5.7963E-02 2.2072E-01 249 8.8078E+00 4.2749E-01 Fluid done Calculating eddy visosity Step 250, t= 8.8393290E+00, DT= 3.1571819E-02, C= 3.334 2.4622E+02 1.0646E+00 Solving for fluid F T T 250 Helmholtz3/fluid: 13 0.3976E-08 0.8000E-08 0.3083E+00 250 U-PRES gmres: 15 8.8405E-08 1.0000E-07 6.2205E-01 6.7180E-02 2.5722E-01 250 8.8393E+00 4.7414E-01 Fluid done Calculating eddy visosity call outfld: ifpsco: F 250 8.8393E+00 Write checkpoint: 0 250 OPEN: poljet3d0.f00011 250 8.8393E+00 done :: Write checkpoint file size = 9.4 MB avg data-throughput = 774.5MB/s io-nodes = 1 call outfld: ifpsco: F 250 8.8393E+00 Write checkpoint: 0 250 OPEN: jet3d0.f00010 250 8.8393E+00 done :: Write checkpoint file size = 9.4 MB avg data-throughput = 847.0MB/s io-nodes = 1 Step 251, t= 8.8645864E+00, DT= 2.5257455E-02, C= 6.125 2.4732E+02 1.1000E+00 Solving for fluid F T T 251 Helmholtz3/fluid: 20 0.3689E-08 0.8000E-08 0.4576E+02 251 U-PRES gmres: 19 7.6441E-08 1.0000E-07 4.7293E+01 8.9832E-02 3.4811E-01 251 8.8646E+00 6.6288E-01 Fluid done Calculating eddy visosity Step 252, t= 8.8847924E+00, DT= 2.0205964E-02, C=304.531 2.4858E+02 1.2546E+00 Solving for fluid F T T 252 201 Unconverged Helmholtz3/Fluid: rbnorm = 0.140100E+18 0.800000E-08 252 U-PRES gmres: 76 9.1868E-08 1.0000E-07 2.8678E+21 3.4409E-01 1.4581E+00 252 8.8848E+00 4.0444E+00 Fluid done Calculating eddy visosity Step 253, t= 8.9009572E+00, DT= 1.6164771E-02, C=******* 2.5326E+02 4.6784E+00 Solving for fluid F T T 253 201 Unconverged Helmholtz3/Fluid: rbnorm = NaN 0.800000E-08 253 U-PRES gmres: 120 NaN 1.0000E-07 NaN 5.4340E-01 2.3310E+00 253 8.9010E+00 4.8603E+00 Fluid done Calculating eddy visosity Step 254, t= 8.9203549E+00, DT= 1.9397725E-02, C=******* 2.5870E+02 5.4484E+00 Solving for fluid F T T 254 201 Unconverged Helmholtz3/Fluid: rbnorm = NaN 0.800000E-08 254 U-PRES gmres: 120 NaN 1.0000E-07 NaN 5.4476E-01 2.3302E+00 254 8.9204E+00 4.8624E+00 Fluid done Calculating eddy visosity Step 255, t= 8.9436322E+00, DT= 2.3277270E-02, C=******* 2.6416E+02 5.4521E+00 Solving for fluid F T T 255 201 Unconverged Helmholtz3/Fluid: rbnorm = NaN 0.800000E-08 255 U-PRES gmres: 120 NaN 1.0000E-07 NaN 5.4309E-01 2.3275E+00 255 8.9436E+00 4.8631E+00 Fluid done Calculating eddy visosity Step 256, t= 8.9715649E+00, DT= 2.7932725E-02, C=******* 2.6961E+02 5.4533E+00 Solving for fluid F T T 256 201 Unconverged Helmholtz3/Fluid: rbnorm = NaN 0.800000E-08 256 U-PRES gmres: 120 NaN 1.0000E-07 NaN 5.8407E-01 2.4309E+00 256 8.9716E+00 4.9608E+00 Fluid done Calculating eddy visosity From nek5000-users at lists.mcs.anl.gov Wed Nov 25 20:53:53 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 26 Nov 2015 10:53:53 +0800 Subject: [Nek5000-users] Rotational velocity Message-ID: Hi neks, Is there a way to specify rotation on a body without mesh deformation? A follow up question to that is..suppose I specify an Omega a value. Omega is rotational velocity. How does nek know that this is rotational and not translational? Rif -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu Nov 26 07:19:23 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 26 Nov 2015 14:19:23 +0100 Subject: [Nek5000-users] Rotational velocity In-Reply-To: References: Message-ID: Hi, if you would like to add Coriolis (and centrifugal) acceleration terms then these can easily be put into userf as body forces. If you use a param-value from the rea file, then the amount of rotation can be set at runtime. If on the other hand you would like to specify a rotating body, then I guess you could use the boundary conditions (userbc) on the body to impose this rotation. Best regards, Philipp On 2015-11-26 03:53, nek5000-users at lists.mcs.anl.gov wrote: > Hi neks, > Is there a way to specify rotation on a body without mesh deformation? A > follow up question to that is..suppose I specify an Omega a value. Omega > is rotational velocity. How does nek know that this is rotational and > not translational? > > > Rif > > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > From nek5000-users at lists.mcs.anl.gov Thu Nov 26 22:34:44 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 26 Nov 2015 23:34:44 -0500 Subject: [Nek5000-users] Nek5000-users Digest, Vol 81, Issue 23 In-Reply-To: References: Message-ID: Leopold, Hi... Interesting LES model; I am also working on a flow where I am adapting the Turb Channel work ... Without analyzing it too deeply, your .usr code looks OK... Not withstanding any problems with the application of the model, have you considered using a shorter time step? The CFL seems kind of high; my runs won't work anywhere near that number ;-); although my code is fully transient and is adding body force for a rotating turbine blade based on time step. When I was working up my code, I did use time steps that would allow runs of several hundred cycles before it blew up. The fact that the program is reducing your DT *could* be an indication that you need a shorter DT... Thanks, Murph -------------------------------------------------------------------------------------------------------------------- Dear Neks, I try to implement the Vreman LES model (see An eddy-viscosity subgrid-scale model for turbulent shear flow: Algebraic theory and applications, A. W. Vreman, Phys. Fl., 16, 10, 2004) to simulate a turbulent jet. To do so, I modified the turbChannel example provided in NEK. However, as soon as I impose in the uservp subroutine a space-varying viscosity, after some time steps, the simulation blows up even if this viscosity remains close from the standard 1/Re viscosity. Has anyone ever experienced this kind of issue? Does anyone have an idea on how to fix this problem? Below are the .usr file and the log of the simulation. Thanks, Best regards, L?opold ------------------------------------------- .usr file ......... ------------------------------------------------------------------------------------------------------------------------ -- Murphy Leo O'Dea PhD Candidate Oakland University mlodea at oakland.edu From nek5000-users at lists.mcs.anl.gov Sat Nov 28 03:13:00 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sat, 28 Nov 2015 17:13:00 +0800 Subject: [Nek5000-users] Nek5000-users Digest, Vol 81, Issue 24 In-Reply-To: References: Message-ID: Hi Philipp, Thanks for the input. Is there a way to make the whole computational domain rotate? Rif On Fri, Nov 27, 2015 at 2:00 AM, wrote: > Send Nek5000-users mailing list submissions to > nek5000-users at lists.mcs.anl.gov > > To subscribe or unsubscribe via the World Wide Web, visit > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > or, via email, send a message with subject or body 'help' to > nek5000-users-request at lists.mcs.anl.gov > > You can reach the person managing the list at > nek5000-users-owner at lists.mcs.anl.gov > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Nek5000-users digest..." > > > Today's Topics: > > 1. Rotational velocity (nek5000-users at lists.mcs.anl.gov) > 2. Re: Rotational velocity (nek5000-users at lists.mcs.anl.gov) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Thu, 26 Nov 2015 10:53:53 +0800 > From: nek5000-users at lists.mcs.anl.gov > To: nek5000-users at lists.mcs.anl.gov > Subject: [Nek5000-users] Rotational velocity > Message-ID: > > Content-Type: text/plain; charset="utf-8" > > Hi neks, > Is there a way to specify rotation on a body without mesh deformation? A > follow up question to that is..suppose I specify an Omega a value. Omega is > rotational velocity. How does nek know that this is rotational and not > translational? > > > Rif > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: < > http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20151126/90f0dfbd/attachment-0001.html > > > > ------------------------------ > > Message: 2 > Date: Thu, 26 Nov 2015 14:19:23 +0100 > From: nek5000-users at lists.mcs.anl.gov > To: nek5000-users at lists.mcs.anl.gov > Subject: Re: [Nek5000-users] Rotational velocity > Message-ID: > > Content-Type: text/plain; charset=windows-1252; format=flowed > > Hi, > if you would like to add Coriolis (and centrifugal) acceleration terms > then these can easily be put into userf as body forces. If you use a > param-value from the rea file, then the amount of rotation can be set at > runtime. > > If on the other hand you would like to specify a rotating body, then I > guess you could use the boundary conditions (userbc) on the body to > impose this rotation. > > Best regards, > Philipp > > On 2015-11-26 03:53, nek5000-users at lists.mcs.anl.gov wrote: > > Hi neks, > > Is there a way to specify rotation on a body without mesh deformation? A > > follow up question to that is..suppose I specify an Omega a value. Omega > > is rotational velocity. How does nek know that this is rotational and > > not translational? > > > > > > Rif > > > > > > _______________________________________________ > > Nek5000-users mailing list > > Nek5000-users at lists.mcs.anl.gov > > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > > > > ------------------------------ > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > > > End of Nek5000-users Digest, Vol 81, Issue 24 > ********************************************* > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Mon Nov 30 06:36:19 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 30 Nov 2015 13:36:19 +0100 Subject: [Nek5000-users] mpi output Message-ID: Hi Neks, I should write a vector of real values in a file. I state that I do not know so much about mpi, Is there an mpi function in Nek that allow me to write it in parallel? sincerely Alessandro -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Mon Nov 30 16:31:13 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 30 Nov 2015 23:31:13 +0100 Subject: [Nek5000-users] Sharing parameters between processors Message-ID: Hi Nekers, In my current simulation, I am defining an array containing parameters that are shared between several processors. I have tried to do this in the usercheck subroutine, with a COMMON statement. However, the COMMON statement enables saving the parameters between two successive calls of the usercheck by the same processor, but does not enable to share a variable between two processor. In other words, if I modify the variable a on the processor 1, the variable a on the processor 2 will not be modified, which is not what I want. Any idea on how to do that? Thanks for your help, Oscar Valdorez -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Mon Nov 30 16:37:26 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 30 Nov 2015 23:37:26 +0100 Subject: [Nek5000-users] Sharing parameters between processors In-Reply-To: References: Message-ID: Indeed, common blocks only work with the local memory. The easiest way is, if it is only a few parameters, that you put them into the rea file in the unused section (say about index 120 or so). Then the values are immediately available on each processor. Otherwise you have to do explicit communication. The starting point would be the broadcast operation, bcast(buf,len), which shares len reals stored on process 0 with all other processes. bcast is a wrapper to the standard MPI function call mpi_bcast (buf,len,mpi_byte,0,nekcomm,ierr) Hope this helps, Philipp On 2015-11-30 23:31, nek5000-users at lists.mcs.anl.gov wrote: > Hi Nekers, > > In my current simulation, I am defining an array containing parameters > that are shared between several processors. I have tried to do this in > the usercheck subroutine, with a COMMON statement. However, the COMMON > statement enables saving the parameters between two successive calls of > the usercheck by the same processor, but does not enable to share a > variable between two processor. In other words, if I modify the variable > a on the processor 1, the variable a on the processor 2 will not be > modified, which is not what I want. Any idea on how to do that? > > Thanks for your help, > > Oscar Valdorez > > > > _______________________________________________ > Nek5000-users mailing list > Nek5000-users at lists.mcs.anl.gov > https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users > From nek5000-users at lists.mcs.anl.gov Mon Nov 30 20:47:03 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 1 Dec 2015 02:47:03 +0000 Subject: [Nek5000-users] Sharing parameters between processors In-Reply-To: References: Message-ID: Hi Oscar, Is there a reason why your quantity is modified only on one processor, and not on the others? How many values are you working with? Paul ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov [nek5000-users-bounces at lists.mcs.anl.gov] on behalf of nek5000-users at lists.mcs.anl.gov [nek5000-users at lists.mcs.anl.gov] Sent: Monday, November 30, 2015 4:31 PM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Sharing parameters between processors Hi Nekers, In my current simulation, I am defining an array containing parameters that are shared between several processors. I have tried to do this in the usercheck subroutine, with a COMMON statement. However, the COMMON statement enables saving the parameters between two successive calls of the usercheck by the same processor, but does not enable to share a variable between two processor. In other words, if I modify the variable a on the processor 1, the variable a on the processor 2 will not be modified, which is not what I want. Any idea on how to do that? Thanks for your help, Oscar Valdorez -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Fri Nov 27 18:40:22 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sat, 28 Nov 2015 01:40:22 +0100 (CET) Subject: [Nek5000-users] mesh problem In-Reply-To: <1233101014.17448452.1448630196035.JavaMail.zimbra@ec-nantes.fr> Message-ID: Dear Neks, I'm trying to simulate the flow on a curved wall with leading edge. My mesh is 130,000 elements and was generated using gambit, then converted using the matlab code Gambittonek. I successfully generated the mesh with genmap (no error or warning), and the calculation is running, however I have discontinuities in my mesh (see attached file). It seems quite independant of reynolds number and element order (checked). Do you have any idea ? I feel like it can come from the partitionning and/or element connectivity, but I can't solve my problem yet. I use around 1300 elements per cpu. Thanks ! Antoine -------------- next part -------------- A non-text attachment was scrubbed... Name: mesh.png Type: image/png Size: 246210 bytes Desc: not available URL: