From nek5000-users at lists.mcs.anl.gov Thu Jan 1 18:08:52 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 1 Jan 2015 19:08:52 -0500 Subject: [Nek5000-users] LES-wall In-Reply-To: References: Message-ID: Hi Tanmoy, I looked into the "turbchannel.usr" example and it seems that the spatial averaging of contractions, i.e., , needs to be modified for other geometries. I Wondered if you have been using eddy-visc() routine for channel geometry or some other geometries. Thanks, Mohsen On 14-12-30 11:29 AM, nek5000-users at lists.mcs.anl.gov wrote: > > Hi Mohsen, > > Yes the turbChannel case is a good point to start with. However, > explicit filtering as LES model only works when using wall-resolved > LES. My past experience shows, to use wall-model, one must have a good > eddy-viscosity model, to properly emulate wall layer physics. Explicit > filtering of LES, is way too bad in preserving the wall physics. > > Sincerely, > Tanmoy > > Best Regards, > Tanmoy > > On Tue, Dec 30, 2014 at 9:07 AM, > wrote: > > Thanks Tanmoy for your suggestion. > > You are right, for a eddy viscosity type of turbulent model, the > Channel example is a good starting point (though, I think it > should be considerably modified to be used in a case with a > different geometry compared to a channel). But I was thinking of > using sort of wall model when employing the explicit filtering of > velocity field as a LES model. > > Thanks, > Mohsen > > On 24-Dec-14, at 11:23 PM, > wrote: > >> Hi Mohsen, >> >> For LES with near wall modelling, you have to develop your own >> script in .usr file. There are no in-built routines except >> eddy_visc() in turbChannel example. If you use eddy-viscosity >> type of near wall models in LES, you can modify the scripts and >> write your own eddy-viscosity routines in uservp. >> >> Thanks, >> Tanmoy >> >> Best Regards, >> Tanmoy >> >> On Thu, Dec 25, 2014 at 12:47 AM, >> > > wrote: >> >> Hello all, >> >> I have a question regarding LES capability of Nek. >> I know that the explicit filtering introduced in Nek can be >> effective for velocity field to provide sort of eddy >> viscosity modeling. Nevertheless, there should be some form >> of modeling for the near wall regions, since it is almost >> infeasible to have DNS like resolution in such regions for >> problems with relatively high Reynolds number and large scale >> geometries. I have seen SFR assemblies example on the Nek web >> in which LES used, however, I am not sure if that was an LES >> with near-wall resolution or with near-wall modeling. >> >> So I was wondering if anyone has any experience of using Nek >> LES with near wall models, or kind of hybrid RANS-LES approaches. >> >> Thanks, >> Mohsen >> >> >> _______________________________________________ >> 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 Thu Jan 1 18:23:11 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 1 Jan 2015 17:23:11 -0700 Subject: [Nek5000-users] LES-wall In-Reply-To: References: Message-ID: Hi Mohsen, Yes you are partially right. The eddy_viscosity routines (dynamic Smagorinsky to be precise) used in "turbChannel.usr" is indeed specific for channel geometries, but they can be applied to geometries having small deviations from channels, like backward facing step, box (lid driven cavity), flat plates with very little modification. They wont work for very complex geometries. It is important to note, that the spatial averaging done in "turbChannel.usr" case is to brute-force spatial homogeneity of the statistics in that direction. If some geometry does not have special homogeneity in one particular direction (e.g. other than periodic boundary conditions) we can comment out the planar average routines. I am personally using 3D flat plate geometry. Best Regards, Tanmoy On Thu, Jan 1, 2015 at 5:08 PM, wrote: > Hi Tanmoy, > > I looked into the "turbchannel.usr" example and it seems that the spatial > averaging of contractions, i.e., , needs to be modified for other > geometries. I Wondered if you have been using eddy-visc() routine for > channel geometry or some other geometries. > > Thanks, > Mohsen > > > On 14-12-30 11:29 AM, nek5000-users at lists.mcs.anl.gov wrote: > > > Hi Mohsen, > > Yes the turbChannel case is a good point to start with. However, explicit > filtering as LES model only works when using wall-resolved LES. My past > experience shows, to use wall-model, one must have a good eddy-viscosity > model, to properly emulate wall layer physics. Explicit filtering of LES, > is way too bad in preserving the wall physics. > > Sincerely, > Tanmoy > > Best Regards, > Tanmoy > > On Tue, Dec 30, 2014 at 9:07 AM, wrote: > >> Thanks Tanmoy for your suggestion. >> >> You are right, for a eddy viscosity type of turbulent model, the Channel >> example is a good starting point (though, I think it should be considerably >> modified to be used in a case with a different geometry compared to a >> channel). But I was thinking of using sort of wall model when employing the >> explicit filtering of velocity field as a LES model. >> >> Thanks, >> Mohsen >> >> On 24-Dec-14, at 11:23 PM, wrote: >> >> Hi Mohsen, >> >> For LES with near wall modelling, you have to develop your own script >> in .usr file. There are no in-built routines except eddy_visc() in >> turbChannel example. If you use eddy-viscosity type of near wall models in >> LES, you can modify the scripts and write your own eddy-viscosity routines >> in uservp. >> >> Thanks, >> Tanmoy >> >> Best Regards, >> Tanmoy >> >> On Thu, Dec 25, 2014 at 12:47 AM, >> wrote: >> >>> Hello all, >>> >>> I have a question regarding LES capability of Nek. >>> I know that the explicit filtering introduced in Nek can be effective >>> for velocity field to provide sort of eddy viscosity modeling. >>> Nevertheless, there should be some form of modeling for the near wall >>> regions, since it is almost infeasible to have DNS like resolution in such >>> regions for problems with relatively high Reynolds number and large scale >>> geometries. I have seen SFR assemblies example on the Nek web in which LES >>> used, however, I am not sure if that was an LES with near-wall resolution >>> or with near-wall modeling. >>> >>> So I was wondering if anyone has any experience of using Nek LES with >>> near wall models, or kind of hybrid RANS-LES approaches. >>> >>> Thanks, >>> Mohsen >>> >>> >>> _______________________________________________ >>> 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 listNek5000-users at lists.mcs.anl.govhttps://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 Thu Jan 1 19:48:54 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 1 Jan 2015 20:48:54 -0500 Subject: [Nek5000-users] LES-wall In-Reply-To: References: Message-ID: Thanks Tanmoy for the information you provided. I am currently working on a problem very like a flow over cylinder, where the cylinder is located in a channel. However, there is no span-wise homogeneity in this problem (as opposed to the flow over 2D cylinder) due to the specific channel shape. Therefore, there is no need to averaging in spanwise planes. Beside planar averaging, however, I couldn't find any thing else in the dynamic SGS model of "turbchannel" example that needs to be modified for my case, as far as I understand. Regards, Mohsen On 15-01-01 07:23 PM, nek5000-users at lists.mcs.anl.gov wrote: > Hi Mohsen, > > Yes you are partially right. The eddy_viscosity routines (dynamic > Smagorinsky to be precise) used in "turbChannel.usr" is indeed > specific for channel geometries, but they can be applied to geometries > having small deviations from channels, like backward facing step, box > (lid driven cavity), flat plates with very little modification. They > wont work for very complex geometries. > > It is important to note, that the spatial averaging done in > "turbChannel.usr" case is to brute-force spatial homogeneity of the > statistics in that direction. If some geometry does not have special > homogeneity in one particular direction (e.g. other than periodic > boundary conditions) we can comment out the planar average routines. > > I am personally using 3D flat plate geometry. > > Best Regards, > Tanmoy > > On Thu, Jan 1, 2015 at 5:08 PM, > wrote: > > Hi Tanmoy, > > I looked into the "turbchannel.usr" example and it seems that the > spatial averaging of contractions, i.e., , needs to be > modified for other geometries. I Wondered if you have been using > eddy-visc() routine for channel geometry or some other geometries. > > Thanks, > Mohsen > > > On 14-12-30 11:29 AM, nek5000-users at lists.mcs.anl.gov > wrote: >> >> Hi Mohsen, >> >> Yes the turbChannel case is a good point to start with. However, >> explicit filtering as LES model only works when using >> wall-resolved LES. My past experience shows, to use wall-model, >> one must have a good eddy-viscosity model, to properly emulate >> wall layer physics. Explicit filtering of LES, is way too bad in >> preserving the wall physics. >> >> Sincerely, >> Tanmoy >> >> Best Regards, >> Tanmoy >> >> On Tue, Dec 30, 2014 at 9:07 AM, > > wrote: >> >> Thanks Tanmoy for your suggestion. >> >> You are right, for a eddy viscosity type of turbulent model, >> the Channel example is a good starting point (though, I think >> it should be considerably modified to be used in a case with >> a different geometry compared to a channel). But I was >> thinking of using sort of wall model when employing the >> explicit filtering of velocity field as a LES model. >> >> Thanks, >> Mohsen >> >> On 24-Dec-14, at 11:23 PM, > > wrote: >> >>> Hi Mohsen, >>> >>> For LES with near wall modelling, you have to develop your >>> own script in .usr file. There are no in-built routines >>> except eddy_visc() in turbChannel example. If you use >>> eddy-viscosity type of near wall models in LES, you can >>> modify the scripts and write your own eddy-viscosity >>> routines in uservp. >>> >>> Thanks, >>> Tanmoy >>> >>> Best Regards, >>> Tanmoy >>> >>> On Thu, Dec 25, 2014 at 12:47 AM, >>> >> > wrote: >>> >>> Hello all, >>> >>> I have a question regarding LES capability of Nek. >>> I know that the explicit filtering introduced in Nek can >>> be effective for velocity field to provide sort of eddy >>> viscosity modeling. Nevertheless, there should be some >>> form of modeling for the near wall regions, since it is >>> almost infeasible to have DNS like resolution in such >>> regions for problems with relatively high Reynolds >>> number and large scale geometries. I have seen SFR >>> assemblies example on the Nek web in which LES used, >>> however, I am not sure if that was an LES with near-wall >>> resolution or with near-wall modeling. >>> >>> So I was wondering if anyone has any experience of using >>> Nek LES with near wall models, or kind of hybrid >>> RANS-LES approaches. >>> >>> Thanks, >>> Mohsen >>> >>> >>> _______________________________________________ >>> 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 > > > _______________________________________________ > 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 Jan 6 04:22:00 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 06 Jan 2015 11:22:00 +0100 Subject: [Nek5000-users] Melting front Message-ID: Dear Neks, I'm interested in the possibility of solving melting front problems with Nek5000 (with Navier-Stokes in the fluid and heat transfer in both fluid and solid). However, I could only find relatively outdated documentation about this topic (User's guide Nekton v2.85 or http://www.mcs.anl.gov/~fischer/Nek5000/nekmanual.pdf for example). Is this option still available in more recent releases of Nek, and should I follow the old documentation to setup my problem or is there a more recent version? Many thanks, Benjamin. From nek5000-users at lists.mcs.anl.gov Tue Jan 6 11:09:08 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Tue, 6 Jan 2015 12:09:08 -0500 Subject: [Nek5000-users] turbchannel-dsavg Message-ID: Hello all, I am working on the Dynamic SGS model used in turbChannel example. There are two geometries in this example, one with 512 elements and the other one with 36 elements. I modified the routine "set_grid_spacing(dg2)" in .usr file to have a non-trivial filter width, rather than a constant one. When using the 36-element case, the code comes to a halt at routine "dofcnt" after calling "userchk", in which the unique gridpoints are counted. I realized that this happens when varaible "dg2" is being averaged on the elements boundaries via "dsavg()" in subroutine "set_grid_spacing(dg2)". However, I couldn't figure out the reason for that. This issue, however, does not happen for 512-element case. So I wondered if anyone has noticed similar issue with this example. I also wanted to echo my previous post (in the following link) on whether or not the SGS model in turbChannel example can be used in other geometries (e.g. for flow over a cylinder). I couldn't find anything that prevents one from using this model in other configurations except the spatial averaging routines, which are used to create a homogeneity in turbulent characteristics over the homogenous planes of channel example. https://lists.mcs.anl.gov/mailman/htdig/nek5000-users/2015-January/003159.html Regards, Mohsen From nek5000-users at lists.mcs.anl.gov Wed Jan 7 18:02:18 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 7 Jan 2015 19:02:18 -0500 Subject: [Nek5000-users] Nek MPI subroutine Message-ID: Hello, I don't understand the following subroutine in the Nek mpi_dummy.f file: --------------------------------------------------------------------------------------------------------- subroutine mpi_comm_rank ( comm, me, ierror ) implicit none integer comm integer ierror integer me integer MPI_FAILURE parameter ( MPI_FAILURE = 1 ) integer MPI_SUCCESS parameter ( MPI_SUCCESS = 0 ) ierror = MPI_SUCCESS me = 0 return end --------------------------------------------------------------------------------------------------------- This will always return one. How does this give the rank of the process then? Can someone please explain the reason for this function? Thanks, Pradeep -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu Jan 8 07:12:49 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 8 Jan 2015 13:12:49 +0000 Subject: [Nek5000-users] Nek MPI subroutine In-Reply-To: References: Message-ID: Pradeep, This file is used only when compiling with mpi turned off. ________________________________ 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, January 07, 2015 6:02 PM To: Nek5000 Subject: [Nek5000-users] Nek MPI subroutine Hello, I don't understand the following subroutine in the Nek mpi_dummy.f file: --------------------------------------------------------------------------------------------------------- subroutine mpi_comm_rank ( comm, me, ierror ) implicit none integer comm integer ierror integer me integer MPI_FAILURE parameter ( MPI_FAILURE = 1 ) integer MPI_SUCCESS parameter ( MPI_SUCCESS = 0 ) ierror = MPI_SUCCESS me = 0 return end --------------------------------------------------------------------------------------------------------- This will always return one. How does this give the rank of the process then? Can someone please explain the reason for this function? Thanks, Pradeep -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Wed Jan 14 02:30:11 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 14 Jan 2015 16:30:11 +0800 Subject: [Nek5000-users] Compilation of nek5000 in ubuntu Message-ID: Hi there, I am compiling the eddy example in my ubuntu OS, but the system returns error with "*Error: Can't open included file 'mpif.h'*". The fortran compiler in makenek file was changed to gfortran since mpif77 gives error: *--------------------------------------------------------------------------** **Unfortunately, this installation of Open MPI was not compiled with** **Fortran 77 support. As such, the mpif77 compiler is non-functional.** **--------------------------------------------------------------------------** **FATAL ERROR: Basic compiler test for mpif77 failed!* Is there anyone experience the same problem when compiling in ubuntu? You assistance will be greatly appreciated. Sincerely, Simon -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Wed Jan 14 03:07:21 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 14 Jan 2015 10:07:21 +0100 Subject: [Nek5000-users] Compilation of nek5000 in ubuntu In-Reply-To: References: Message-ID: Hi Simon, I would suggest you to reinstall OPEN-MPI in ubuntu and use mpif77 compiler. Best, Kamal PS: Install Open Mpi in a separate folder and the add the path in ./bashsrc or to the makenek directly On 14/01/2015 09:30, nek5000-users at lists.mcs.anl.gov wrote: > Hi there, > > I am compiling the eddy example in my ubuntu OS, but the system > returns error with "*Error: Can't open included file 'mpif.h'*". The > fortran compiler in makenek file was changed to gfortran since mpif77 > gives error: > *--------------------------------------------------------------------------** > **Unfortunately, this installation of Open MPI was not compiled with** > **Fortran 77 support. As such, the mpif77 compiler is non-functional.** > **--------------------------------------------------------------------------** > **FATAL ERROR: Basic compiler test for mpif77 failed!* > > Is there anyone experience the same problem when compiling in ubuntu? > You assistance will be greatly appreciated. > > Sincerely, > Simon > > > _______________________________________________ > 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 Wed Jan 14 03:24:10 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 14 Jan 2015 09:24:10 +0000 Subject: [Nek5000-users] Compilation of nek5000 in ubuntu In-Reply-To: References: , Message-ID: Hi Kamal, Thank you very much for the suggestion. As I am a Linux user at very basic level, could you please tell a little bit more details on reinstalling open-mpi? Are u suggesting me reinstall the same version as installed in ubundu, or any other version? If it should be a certain version, I will download and install in. To tell the truth, I installed a package f2c that can translate fortran77 to c language, but it does not help here. Do you think do u have to install fortran77 in Ubuntu? It is said installing f77 will cause more errors in the latest version of Ubuntu. Thanks for your reply again. Sincerely, Simon On 14 Jan, 2015, at 5:08 pm, "nek5000-users at lists.mcs.anl.gov" > wrote: Hi Simon, I would suggest you to reinstall OPEN-MPI in ubuntu and use mpif77 compiler. Best, Kamal PS: Install Open Mpi in a separate folder and the add the path in ./bashsrc or to the makenek directly On 14/01/2015 09:30, nek5000-users at lists.mcs.anl.gov wrote: Hi there, I am compiling the eddy example in my ubuntu OS, but the system returns error with "Error: Can't open included file 'mpif.h'". The fortran compiler in makenek file was changed to gfortran since mpif77 gives error: -------------------------------------------------------------------------- Unfortunately, this installation of Open MPI was not compiled with Fortran 77 support. As such, the mpif77 compiler is non-functional. -------------------------------------------------------------------------- FATAL ERROR: Basic compiler test for mpif77 failed! Is there anyone experience the same problem when compiling in ubuntu? You assistance will be greatly appreciated. Sincerely, Simon _______________________________________________ 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 Wed Jan 14 03:58:26 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 14 Jan 2015 10:58:26 +0100 Subject: [Nek5000-users] Compilation of nek5000 in ubuntu In-Reply-To: References: , Message-ID: Hi Simon, 1) Download the Open-Mpi file from here : www.open-mpi.org/software/ompi/v1.8/downloads/openmpi-1.8.4.tar.gz, unpack it 2) Create a folder ' sudo mkdir /usr/local/openmpi ' 3) Go to the unpacked openmpi-1.8.4 folder and do ./configure --prefix = /usr/local/openmpi 4) Then just type ' make ' 5) Then ' make check install ' This will install open-mpi in the folder /usr/local/openmpi, now your mpi files will be in this folder. 6) Open makenek and change ' F77 = /use/local/openmpi/mpif77 ' and ' CC = /usr/local/openmpi/mpicc ' or you can the path to ./bashrc file if you don't want to do this step. I hope this should work for you. Best Regards, Kamal On 14/01/2015 10:24, nek5000-users at lists.mcs.anl.gov wrote: > Hi Kamal, > > Thank you very much for the suggestion. As I am a Linux user at very > basic level, could you please tell a little bit more details on > reinstalling open-mpi? Are u suggesting me reinstall the same version > as installed in ubundu, or any other version? If it should be a > certain version, I will download and install in. > > To tell the truth, I installed a package f2c that can translate > fortran77 to c language, but it does not help here. Do you think do u > have to install fortran77 in Ubuntu? It is said installing f77 will > cause more errors in the latest version of Ubuntu. > > Thanks for your reply again. > > Sincerely, > Simon > > On 14 Jan, 2015, at 5:08 pm, "nek5000-users at lists.mcs.anl.gov > " > > wrote: > >> Hi Simon, >> >> I would suggest you to reinstall OPEN-MPI in ubuntu and use mpif77 >> compiler. >> >> Best, >> Kamal >> >> PS: Install Open Mpi in a separate folder and the add the path in >> ./bashsrc or to the makenek directly >> >> >> On 14/01/2015 09:30, nek5000-users at lists.mcs.anl.gov wrote: >>> Hi there, >>> >>> I am compiling the eddy example in my ubuntu OS, but the system >>> returns error with "*Error: Can't open included file 'mpif.h'*". The >>> fortran compiler in makenek file was changed to gfortran since >>> mpif77 gives error: >>> *--------------------------------------------------------------------------** >>> **Unfortunately, this installation of Open MPI was not compiled with** >>> **Fortran 77 support. As such, the mpif77 compiler is non-functional.** >>> **--------------------------------------------------------------------------** >>> **FATAL ERROR: Basic compiler test for mpif77 failed!* >>> >>> Is there anyone experience the same problem when compiling in >>> ubuntu? You assistance will be greatly appreciated. >>> >>> Sincerely, >>> Simon >>> >>> >>> _______________________________________________ >>> 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 Wed Jan 14 05:31:37 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 14 Jan 2015 11:31:37 +0000 Subject: [Nek5000-users] Compilation of nek5000 in ubuntu In-Reply-To: References: , , Message-ID: Hi Kamal, I am really thankful that you give so detailed information on compilation of Open-mpi. Everything went well when following your first 4 steps except for Step 5 in which creation of dir /sur/local/openmpi/share was denied. So I re-do Step 5 with command "sudo make check install", and then it compiled well without errors. Then I followed Step 6 but it was found that no mpif77 nor mpicc files were under dir /use/local/openmpi/. After checking files under dir /use/local/openmpi/bin, there exist mpiff77 and mpicc files. So finally I changed makenek to "F77 = /usr/local/openmpi/bin/mpif77" and " CC = /usr/local/openmpi/bin/mpicc" and then type "./makenek eddy_uv". Unluckily, the error came again as following: -------------------------------------------------------------------------- Unfortunately, this installation of Open MPI was not compiled with Fortran 77 support. As such, the mpif77 compiler is non-functional. -------------------------------------------------------------------------- FATAL ERROR: Basic compiler test for /usr/local/openmpi/bin/mpif77 failed! It seems fortran 77 not ready yet. Do you have any comments about it? Thanks, Simon ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov on behalf of nek5000-users at lists.mcs.anl.gov Sent: Wednesday, 14 January, 2015 5:58 PM To: nek5000-users at lists.mcs.anl.gov Subject: Re: [Nek5000-users] Compilation of nek5000 in ubuntu Hi Simon, 1) Download the Open-Mpi file from here : www.open-mpi.org/software/ompi/v1.8/downloads/openmpi-1.8.4.tar.gz, unpack it 2) Create a folder ' sudo mkdir /usr/local/openmpi ' 3) Go to the unpacked openmpi-1.8.4 folder and do ./configure --prefix = /usr/local/openmpi 4) Then just type ' make ' 5) Then ' make check install ' This will install open-mpi in the folder /usr/local/openmpi, now your mpi files will be in this folder. 6) Open makenek and change ' F77 = /use/local/openmpi/mpif77 ' and ' CC = /usr/local/openmpi/mpicc ' or you can the path to ./bashrc file if you don't want to do this step. I hope this should work for you. Best Regards, Kamal On 14/01/2015 10:24, nek5000-users at lists.mcs.anl.gov wrote: Hi Kamal, Thank you very much for the suggestion. As I am a Linux user at very basic level, could you please tell a little bit more details on reinstalling open-mpi? Are u suggesting me reinstall the same version as installed in ubundu, or any other version? If it should be a certain version, I will download and install in. To tell the truth, I installed a package f2c that can translate fortran77 to c language, but it does not help here. Do you think do u have to install fortran77 in Ubuntu? It is said installing f77 will cause more errors in the latest version of Ubuntu. Thanks for your reply again. Sincerely, Simon On 14 Jan, 2015, at 5:08 pm, "nek5000-users at lists.mcs.anl.gov" > wrote: Hi Simon, I would suggest you to reinstall OPEN-MPI in ubuntu and use mpif77 compiler. Best, Kamal PS: Install Open Mpi in a separate folder and the add the path in ./bashsrc or to the makenek directly On 14/01/2015 09:30, nek5000-users at lists.mcs.anl.gov wrote: Hi there, I am compiling the eddy example in my ubuntu OS, but the system returns error with "Error: Can't open included file 'mpif.h'". The fortran compiler in makenek file was changed to gfortran since mpif77 gives error: -------------------------------------------------------------------------- Unfortunately, this installation of Open MPI was not compiled with Fortran 77 support. As such, the mpif77 compiler is non-functional. -------------------------------------------------------------------------- FATAL ERROR: Basic compiler test for mpif77 failed! Is there anyone experience the same problem when compiling in ubuntu? You assistance will be greatly appreciated. Sincerely, Simon _______________________________________________ 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 Wed Jan 14 05:49:19 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 14 Jan 2015 12:49:19 +0100 Subject: [Nek5000-users] Compilation of nek5000 in ubuntu In-Reply-To: References: , , Message-ID: Hi Simon, Could you try this ' export LD_LIBRARY_PATH=/usr/local/opempi/lib ' and then launching "./makenek eddy_uv". If that does not help, There are a couple of things to do : 1) sudo apt-get update 2) sudo apt-get upgrade 3) Sudo apt-get install build essential do the above commands to install some Linux libraries and try recompiling openmpi with ' ./configure --prefix=/usr/local/openmpi FC=/usr/local/gfortran (path of your gfortran). just take look at the output when you do ' sudo make ' and see If everything is going perfectly. Best, Kamal PS: Check If your gfortran works properly On 14/01/2015 12:31, nek5000-users at lists.mcs.anl.gov wrote: > > Hi Kamal, > > > I am really thankful that you give so detailed information on > compilation of Open-mpi. Everything went well when following your > first 4 steps except for Step 5 in which creation of dir > /sur/local/openmpi/share was denied. So I re-do Step 5 with command > "sudo make check install", and then it compiled well without errors. > > > Then I followed Step 6 but it was found that no mpif77 nor mpicc files > were under dir /use/local/openmpi/. After checking files under dir > /use/local/openmpi/bin, there exist mpiff77 and mpicc files. So > finally I changed makenek to "F77 = /usr/local/openmpi/bin/mpif77" and > " CC = /usr/local/openmpi/bin/mpicc" and then type "./makenek eddy_uv". > > > Unluckily, the error came again as following: > > -------------------------------------------------------------------------- > Unfortunately, this installation of Open MPI was not compiled with > Fortran 77 support. As such, the mpif77 compiler is non-functional. > -------------------------------------------------------------------------- > FATAL ERROR: Basic compiler test for /usr/local/openmpi/bin/mpif77 failed! > > > It seems fortran 77 not ready yet. Do you have any comments about it? > > > Thanks, > > Simon > > > ------------------------------------------------------------------------ > *From:* nek5000-users-bounces at lists.mcs.anl.gov > on behalf of > nek5000-users at lists.mcs.anl.gov > *Sent:* Wednesday, 14 January, 2015 5:58 PM > *To:* nek5000-users at lists.mcs.anl.gov > *Subject:* Re: [Nek5000-users] Compilation of nek5000 in ubuntu > Hi Simon, > > 1) Download the Open-Mpi file from here : > www.open-mpi.org/software/ompi/v1.8/downloads/openmpi-1.8.4.tar.gz, > unpack it > 2) Create a folder ' sudo mkdir /usr/local/openmpi ' > 3) Go to the unpacked openmpi-1.8.4 folder and do ./configure > --prefix = /usr/local/openmpi > 4) Then just type ' make ' > 5) Then ' make check install ' > > This will install open-mpi in the folder /usr/local/openmpi, now > your mpi files will be in this folder. > > 6) Open makenek and change ' F77 = /use/local/openmpi/mpif77 ' and > ' CC = /usr/local/openmpi/mpicc ' or you can the path to ./bashrc > file if you don't want to do this step. > > I hope this should work for you. > > > Best Regards, > Kamal > > > > > On 14/01/2015 10:24, nek5000-users at lists.mcs.anl.gov wrote: >> Hi Kamal, >> >> Thank you very much for the suggestion. As I am a Linux user at very >> basic level, could you please tell a little bit more details on >> reinstalling open-mpi? Are u suggesting me reinstall the same version >> as installed in ubundu, or any other version? If it should be a >> certain version, I will download and install in. >> >> To tell the truth, I installed a package f2c that can translate >> fortran77 to c language, but it does not help here. Do you think do u >> have to install fortran77 in Ubuntu? It is said installing f77 will >> cause more errors in the latest version of Ubuntu. >> >> Thanks for your reply again. >> >> Sincerely, >> Simon >> >> On 14 Jan, 2015, at 5:08 pm, "nek5000-users at lists.mcs.anl.gov >> " >> > > wrote: >> >>> Hi Simon, >>> >>> I would suggest you to reinstall OPEN-MPI in ubuntu and use mpif77 >>> compiler. >>> >>> Best, >>> Kamal >>> >>> PS: Install Open Mpi in a separate folder and the add the path in >>> ./bashsrc or to the makenek directly >>> >>> >>> On 14/01/2015 09:30, nek5000-users at lists.mcs.anl.gov wrote: >>>> Hi there, >>>> >>>> I am compiling the eddy example in my ubuntu OS, but the system >>>> returns error with "*Error: Can't open included file 'mpif.h'*". >>>> The fortran compiler in makenek file was changed to gfortran since >>>> mpif77 gives error: >>>> *--------------------------------------------------------------------------** >>>> **Unfortunately, this installation of Open MPI was not compiled with** >>>> **Fortran 77 support. As such, the mpif77 compiler is >>>> non-functional.** >>>> **--------------------------------------------------------------------------** >>>> **FATAL ERROR: Basic compiler test for mpif77 failed!* >>>> >>>> Is there anyone experience the same problem when compiling in >>>> ubuntu? You assistance will be greatly appreciated. >>>> >>>> Sincerely, >>>> Simon >>>> >>>> >>>> _______________________________________________ >>>> 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 Wed Jan 14 07:17:06 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 14 Jan 2015 13:17:06 +0000 Subject: [Nek5000-users] Compilation of nek5000 in ubuntu In-Reply-To: References: , , , Message-ID: Hi Kamal, Unfortunately, your suggestions does not change much of the compiling. The mpif77 compiler is still non-functional. When I do "sudo make" or "sudo make check install", no errors or warning appears except repeated similar words "Nothing to be done for 'xxxx (eg. all)' ". And when I do "sudo apt-get install build essential", it actually did not find out essential. I will re-do the steps tomorrow to double check. Thanks again for your assistance. Regards, Simon ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov on behalf of nek5000-users at lists.mcs.anl.gov Sent: Wednesday, 14 January, 2015 7:49 PM To: nek5000-users at lists.mcs.anl.gov Subject: Re: [Nek5000-users] Compilation of nek5000 in ubuntu Hi Simon, Could you try this ' export LD_LIBRARY_PATH=/usr/local/opempi/lib ' and then launching "./makenek eddy_uv". If that does not help, There are a couple of things to do : 1) sudo apt-get update 2) sudo apt-get upgrade 3) Sudo apt-get install build essential do the above commands to install some Linux libraries and try recompiling openmpi with ' ./configure --prefix=/usr/local/openmpi FC=/usr/local/gfortran (path of your gfortran). just take look at the output when you do ' sudo make ' and see If everything is going perfectly. Best, Kamal PS: Check If your gfortran works properly On 14/01/2015 12:31, nek5000-users at lists.mcs.anl.gov wrote: Hi Kamal, I am really thankful that you give so detailed information on compilation of Open-mpi. Everything went well when following your first 4 steps except for Step 5 in which creation of dir /sur/local/openmpi/share was denied. So I re-do Step 5 with command "sudo make check install", and then it compiled well without errors. Then I followed Step 6 but it was found that no mpif77 nor mpicc files were under dir /use/local/openmpi/. After checking files under dir /use/local/openmpi/bin, there exist mpiff77 and mpicc files. So finally I changed makenek to "F77 = /usr/local/openmpi/bin/mpif77" and " CC = /usr/local/openmpi/bin/mpicc" and then type "./makenek eddy_uv". Unluckily, the error came again as following: -------------------------------------------------------------------------- Unfortunately, this installation of Open MPI was not compiled with Fortran 77 support. As such, the mpif77 compiler is non-functional. -------------------------------------------------------------------------- FATAL ERROR: Basic compiler test for /usr/local/openmpi/bin/mpif77 failed! It seems fortran 77 not ready yet. Do you have any comments about it? Thanks, Simon ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov on behalf of nek5000-users at lists.mcs.anl.gov Sent: Wednesday, 14 January, 2015 5:58 PM To: nek5000-users at lists.mcs.anl.gov Subject: Re: [Nek5000-users] Compilation of nek5000 in ubuntu Hi Simon, 1) Download the Open-Mpi file from here : www.open-mpi.org/software/ompi/v1.8/downloads/openmpi-1.8.4.tar.gz, unpack it 2) Create a folder ' sudo mkdir /usr/local/openmpi ' 3) Go to the unpacked openmpi-1.8.4 folder and do ./configure --prefix = /usr/local/openmpi 4) Then just type ' make ' 5) Then ' make check install ' This will install open-mpi in the folder /usr/local/openmpi, now your mpi files will be in this folder. 6) Open makenek and change ' F77 = /use/local/openmpi/mpif77 ' and ' CC = /usr/local/openmpi/mpicc ' or you can the path to ./bashrc file if you don't want to do this step. I hope this should work for you. Best Regards, Kamal On 14/01/2015 10:24, nek5000-users at lists.mcs.anl.gov wrote: Hi Kamal, Thank you very much for the suggestion. As I am a Linux user at very basic level, could you please tell a little bit more details on reinstalling open-mpi? Are u suggesting me reinstall the same version as installed in ubundu, or any other version? If it should be a certain version, I will download and install in. To tell the truth, I installed a package f2c that can translate fortran77 to c language, but it does not help here. Do you think do u have to install fortran77 in Ubuntu? It is said installing f77 will cause more errors in the latest version of Ubuntu. Thanks for your reply again. Sincerely, Simon On 14 Jan, 2015, at 5:08 pm, "nek5000-users at lists.mcs.anl.gov" > wrote: Hi Simon, I would suggest you to reinstall OPEN-MPI in ubuntu and use mpif77 compiler. Best, Kamal PS: Install Open Mpi in a separate folder and the add the path in ./bashsrc or to the makenek directly On 14/01/2015 09:30, nek5000-users at lists.mcs.anl.gov wrote: Hi there, I am compiling the eddy example in my ubuntu OS, but the system returns error with "Error: Can't open included file 'mpif.h'". The fortran compiler in makenek file was changed to gfortran since mpif77 gives error: -------------------------------------------------------------------------- Unfortunately, this installation of Open MPI was not compiled with Fortran 77 support. As such, the mpif77 compiler is non-functional. -------------------------------------------------------------------------- FATAL ERROR: Basic compiler test for mpif77 failed! Is there anyone experience the same problem when compiling in ubuntu? You assistance will be greatly appreciated. Sincerely, Simon _______________________________________________ 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 Wed Jan 14 21:57:53 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 14 Jan 2015 22:57:53 -0500 Subject: [Nek5000-users] Parallel on several nodes Message-ID: Hi guys, I have a naive questions but I couldn't find my answer in existing posts. Is it possible to run Nek on several nodes, say, on a cluster? If so, which script should be used? A command like /nekmpi eddy_uv 4/ does not help becaue on each node I have 8 CPUs but I'd like to have my 16 nodes (each have 12 CPUs) run together. Any help is appreciated. Saleh From nek5000-users at lists.mcs.anl.gov Wed Jan 14 22:06:51 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 14 Jan 2015 21:06:51 -0700 Subject: [Nek5000-users] Parallel on several nodes In-Reply-To: References: Message-ID: Hi Saleh, You can develop a shell script (for example a PBS script for your cluster), a .sh file, that you can submit to the queue. The .sh file would be very similar to the nekbmpi script, with some extra headers, that will specify, the number of nodes, type of job, wall clock time etc. ************************************************************************************************************ #! /bin/bash #PBS -N turbChannel #PBS -l walltime=400:00:00 #PBS -l nodes=6:ppn=16 #PBS -j oe #PBS -o sparkyLog cd $PBS_O_WORKDIR jobName=turbChannel #name of the *.rea file nprocs=96 #number of processors echo $jobName > SESSION.NAME echo `pwd`'/' >> SESSION.NAME touch $jobName.rea rm -f logfile rm -f ioinfo mv $jobName.log.$nprocs $jobName.log1.$nprocs mv $jobName.sch $jobName.sch1 mpiexec ./nek5000 > $jobName.log.$nprocs sleep 2 ln $jobName.log.$nprocs logfile rm -f SESSION.NAME ************************************************************************************************************** You can see a PBS script that I use to run the job; the script name staring from jobname, is just the script of nekbmpi with the running variable names replaced by specific numeric and strings (jobname, number of processors etc.) HtH, Tanmoy Best Regards, Tanmoy On Wed, Jan 14, 2015 at 8:57 PM, wrote: > Hi guys, > > I have a naive questions but I couldn't find my answer in existing posts. > > Is it possible to run Nek on several nodes, say, on a cluster? If so, > which script should be used? > > A command like /nekmpi eddy_uv 4/ does not help becaue on each node I > have 8 CPUs but I'd like to have my 16 nodes (each have 12 CPUs) run > together. > > Any help is appreciated. > > Saleh > > > _______________________________________________ > 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 Jan 15 04:33:58 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 15 Jan 2015 11:33:58 +0100 Subject: [Nek5000-users] Compilation of nek5000 in ubuntu Message-ID: Hi, Simon I met the similar problem as you before. What I did is : 1) check which and where mpif77 have been installed sth at sth:~$ which mpif77 /usr/.../mpif77 2) change this path to the makenek F77="/usr/.../mpif77" also you can include this path to you profile then you don't need to change makenek every time. hope it helps Qiong -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu Jan 15 06:15:51 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 15 Jan 2015 12:15:51 +0000 Subject: [Nek5000-users] Compilation of nek5000 in ubuntu In-Reply-To: References: Message-ID: Hi Qiong, Thanks for your help. By checking mpif77, the location of mpif77 is: /home/szhang009/OpenFOAM/ThirdParty-2.3.0/platforms/linux64Gcc/openmpi-1.6.5/bin/mpif77 which is in the OpenFOAM file. So I copy and paste it to makenek but it does not work where it still says mpif77 compiler is non-functional. I also checked the dir /usr/bin and there also exists mpif77. So I change makenek to this address but it still returns the same error. Thanks & Regards, Simon ________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov on behalf of nek5000-users at lists.mcs.anl.gov Sent: Thursday, 15 January, 2015 6:33 PM To: nek5000-users at lists.mcs.anl.gov Subject: Re: [Nek5000-users] Compilation of nek5000 in ubuntu Hi, Simon I met the similar problem as you before. What I did is : 1) check which and where mpif77 have been installed sth at sth:~$ which mpif77 /usr/.../mpif77 2) change this path to the makenek F77="/usr/.../mpif77" also you can include this path to you profile then you don't need to change makenek every time. hope it helps Qiong -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu Jan 15 12:16:14 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 15 Jan 2015 13:16:14 -0500 Subject: [Nek5000-users] Parallel on several nodes In-Reply-To: References: Message-ID: Hi Tanmoy, Many thanks for prompt and very useful response. On 1/14/2015 11:06 PM, nek5000-users at lists.mcs.anl.gov wrote: > Hi Saleh, > > You can develop a shell script (for example a PBS script for your > cluster), a .sh file, that you can submit to the queue. The .sh file > would be very similar to the nekbmpi script, with some extra headers, > that will specify, the number of nodes, type of job, wall clock time etc. > > ************************************************************************************************************ > #! /bin/bash > > #PBS -N turbChannel > #PBS -l walltime=400:00:00 > #PBS -l nodes=6:ppn=16 > #PBS -j oe > #PBS -o sparkyLog > > cd $PBS_O_WORKDIR > > > jobName=turbChannel #name of the *.rea file > nprocs=96 #number of processors > > echo $jobName > SESSION.NAME > echo `pwd`'/' >> SESSION.NAME > touch $jobName.rea > rm -f logfile > rm -f ioinfo > mv $jobName.log.$nprocs $jobName.log1.$nprocs > mv $jobName.sch $jobName.sch1 > mpiexec ./nek5000 > $jobName.log.$nprocs > sleep 2 > ln $jobName.log.$nprocs logfile > rm -f SESSION.NAME > > ************************************************************************************************************** > > You can see a PBS script that I use to run the job; the script name > staring from jobname, is just the script of nekbmpi with the running > variable names replaced by specific numeric and strings (jobname, > number of processors etc.) > > HtH, > Tanmoy > > Best Regards, > Tanmoy > > On Wed, Jan 14, 2015 at 8:57 PM, > wrote: > > Hi guys, > > I have a naive questions but I couldn't find my answer in existing > posts. > > Is it possible to run Nek on several nodes, say, on a cluster? If > so, which script should be used? > > A command like /nekmpi eddy_uv 4/ does not help becaue on each > node I have 8 CPUs but I'd like to have my 16 nodes (each have 12 > CPUs) run together. > > Any help is appreciated. > > Saleh > > > _______________________________________________ > 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 Wed Jan 21 03:53:00 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 21 Jan 2015 12:53:00 +0300 Subject: [Nek5000-users] bc for a jet in stratified fluid Message-ID: An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Wed Jan 21 15:53:36 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 21 Jan 2015 21:53:36 +0000 Subject: [Nek5000-users] Mesh-morphing for bend pipe Message-ID: Dear Neks, I have been looking at the mesh-morphing for a bend pipe recently. I found the thread https://lists.mcs.anl.gov/mailman/htdig/nek5000-users/2011-June/001398.html on how to bend a straight pipe. The code worked perfectly for bending a straight pipe into a 90 degree bend shape. The problem is I need to bend my pipe at certain section, say I have a pipe with total length of 10 and I want to bend it in the second half, i.e. from 5-10. However, when I tried to do it in this way, the error of 'Vanishing Jacobian near Xth node of element XX' appeared and the simulation couldn't start. I can't see why it doesn't work. Below is the code I'm using to bend the pipe (from 5-10) into a 90 degree bend. Hope someone could help me on this. Any suggestion would be appreciated. Thank you very much in advance. Kind regards, Tony c----------------------------------------------------------------------- subroutine usrdat2() ! This routine to modify mesh coordinates include 'SIZE' include 'TOTAL' n = nx1*ny1*nz1*nelv c First, rotate x into axial position do i=1,n x_original = xm1(i,1,1,1) y_original = ym1(i,1,1,1) z_original = zm1(i,1,1,1) xm1(i,1,1,1) = z_original ym1(i,1,1,1) = y_original zm1(i,1,1,1) = -x_original enddo c Second, bend pipe into 90 degree bend xmin = glmin(xm1,n) xmax = glmax(xm1,n) ymin = glmin(ym1,n) ymax = glmax(ym1,n) if (nid==0) print *,xmin,xmax,ymin,ymax !which are 0, 10, -1, 1 rad0 = 0.5*(ymax-ymin) ! Radius of initial pipe rad1 = 1.0 ! Radius of new pipe radm = 2.0 ! Major radius of torus do i=1,n x = xm1(i,1,1,1) y = ym1(i,1,1,1) if (x .gt. 5.0e0) then theta = 0.5*pi*(x-5.0e0)/(xmax-5.0e0) rad = radm + 2.0e0*rad1*(y-ymin)/(ymax-ymin) ym1(i,1,1,1) = rad*cos(theta) xm1(i,1,1,1) = rad*sin(theta)+5.0e0 end if end do return end -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Wed Jan 21 16:51:23 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Wed, 21 Jan 2015 22:51:23 +0000 Subject: [Nek5000-users] Mesh-morphing for bend pipe In-Reply-To: References: Message-ID: Hi Tony, It appears that you shift y, but only for x > 5, so there is a sudden jump in the y value at x=5. Another possible thing that I didn't check is to make certain that your Rcos theta and R sin theta transformation preserves the right-hand-rule. (Your first rotation to x=axial direction does so.) Often the fastest way to identify these issues is to run a tiny case on one processor and just dump out all points (x,y,z) and plot them in matlab or via splot in gnuplot. hth, 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, January 21, 2015 3:53 PM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Mesh-morphing for bend pipe Dear Neks, I have been looking at the mesh-morphing for a bend pipe recently. I found the thread https://lists.mcs.anl.gov/mailman/htdig/nek5000-users/2011-June/001398.html on how to bend a straight pipe. The code worked perfectly for bending a straight pipe into a 90 degree bend shape. The problem is I need to bend my pipe at certain section, say I have a pipe with total length of 10 and I want to bend it in the second half, i.e. from 5-10. However, when I tried to do it in this way, the error of 'Vanishing Jacobian near Xth node of element XX' appeared and the simulation couldn't start. I can't see why it doesn't work. Below is the code I'm using to bend the pipe (from 5-10) into a 90 degree bend. Hope someone could help me on this. Any suggestion would be appreciated. Thank you very much in advance. Kind regards, Tony c----------------------------------------------------------------------- subroutine usrdat2() ! This routine to modify mesh coordinates include 'SIZE' include 'TOTAL' n = nx1*ny1*nz1*nelv c First, rotate x into axial position do i=1,n x_original = xm1(i,1,1,1) y_original = ym1(i,1,1,1) z_original = zm1(i,1,1,1) xm1(i,1,1,1) = z_original ym1(i,1,1,1) = y_original zm1(i,1,1,1) = -x_original enddo c Second, bend pipe into 90 degree bend xmin = glmin(xm1,n) xmax = glmax(xm1,n) ymin = glmin(ym1,n) ymax = glmax(ym1,n) if (nid==0) print *,xmin,xmax,ymin,ymax !which are 0, 10, -1, 1 rad0 = 0.5*(ymax-ymin) ! Radius of initial pipe rad1 = 1.0 ! Radius of new pipe radm = 2.0 ! Major radius of torus do i=1,n x = xm1(i,1,1,1) y = ym1(i,1,1,1) if (x .gt. 5.0e0) then theta = 0.5*pi*(x-5.0e0)/(xmax-5.0e0) rad = radm + 2.0e0*rad1*(y-ymin)/(ymax-ymin) ym1(i,1,1,1) = rad*cos(theta) xm1(i,1,1,1) = rad*sin(theta)+5.0e0 end if end do return end -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu Jan 22 20:37:53 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 23 Jan 2015 02:37:53 +0000 Subject: [Nek5000-users] Nek5000-users Digest, Vol 71, Issue 11 In-Reply-To: References: Message-ID: Dear Paul, Thank you very much for your help. Yes, the problem was the jump in y direction. I wasn't aware of the shift in y direction after bending the pipe. The code works fine now. Just a quick question, if I want to do a U-bend pipe simulation, should I just change my geometry in usrdat2 subroutine or are there any other ways you would suggest? Thanks again. Cheers, Tony ________________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov on behalf of nek5000-users-request at lists.mcs.anl.gov Sent: 22 January 2015 18:00 To: nek5000-users at lists.mcs.anl.gov Subject: Nek5000-users Digest, Vol 71, Issue 11 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. Mesh-morphing for bend pipe (nek5000-users at lists.mcs.anl.gov) 2. Re: Mesh-morphing for bend pipe (nek5000-users at lists.mcs.anl.gov) ---------------------------------------------------------------------- Message: 1 Date: Wed, 21 Jan 2015 21:53:36 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: [Nek5000-users] Mesh-morphing for bend pipe Message-ID: Content-Type: text/plain; charset="iso-8859-1" Dear Neks, I have been looking at the mesh-morphing for a bend pipe recently. I found the thread https://lists.mcs.anl.gov/mailman/htdig/nek5000-users/2011-June/001398.html on how to bend a straight pipe. The code worked perfectly for bending a straight pipe into a 90 degree bend shape. The problem is I need to bend my pipe at certain section, say I have a pipe with total length of 10 and I want to bend it in the second half, i.e. from 5-10. However, when I tried to do it in this way, the error of 'Vanishing Jacobian near Xth node of element XX' appeared and the simulation couldn't start. I can't see why it doesn't work. Below is the code I'm using to bend the pipe (from 5-10) into a 90 degree bend. Hope someone could help me on this. Any suggestion would be appreciated. Thank you very much in advance. Kind regards, Tony c----------------------------------------------------------------------- subroutine usrdat2() ! This routine to modify mesh coordinates include 'SIZE' include 'TOTAL' n = nx1*ny1*nz1*nelv c First, rotate x into axial position do i=1,n x_original = xm1(i,1,1,1) y_original = ym1(i,1,1,1) z_original = zm1(i,1,1,1) xm1(i,1,1,1) = z_original ym1(i,1,1,1) = y_original zm1(i,1,1,1) = -x_original enddo c Second, bend pipe into 90 degree bend xmin = glmin(xm1,n) xmax = glmax(xm1,n) ymin = glmin(ym1,n) ymax = glmax(ym1,n) if (nid==0) print *,xmin,xmax,ymin,ymax !which are 0, 10, -1, 1 rad0 = 0.5*(ymax-ymin) ! Radius of initial pipe rad1 = 1.0 ! Radius of new pipe radm = 2.0 ! Major radius of torus do i=1,n x = xm1(i,1,1,1) y = ym1(i,1,1,1) if (x .gt. 5.0e0) then theta = 0.5*pi*(x-5.0e0)/(xmax-5.0e0) rad = radm + 2.0e0*rad1*(y-ymin)/(ymax-ymin) ym1(i,1,1,1) = rad*cos(theta) xm1(i,1,1,1) = rad*sin(theta)+5.0e0 end if end do return end -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Message: 2 Date: Wed, 21 Jan 2015 22:51:23 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: Re: [Nek5000-users] Mesh-morphing for bend pipe Message-ID: Content-Type: text/plain; charset="iso-8859-1" Hi Tony, It appears that you shift y, but only for x > 5, so there is a sudden jump in the y value at x=5. Another possible thing that I didn't check is to make certain that your Rcos theta and R sin theta transformation preserves the right-hand-rule. (Your first rotation to x=axial direction does so.) Often the fastest way to identify these issues is to run a tiny case on one processor and just dump out all points (x,y,z) and plot them in matlab or via splot in gnuplot. hth, 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, January 21, 2015 3:53 PM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Mesh-morphing for bend pipe Dear Neks, I have been looking at the mesh-morphing for a bend pipe recently. I found the thread https://lists.mcs.anl.gov/mailman/htdig/nek5000-users/2011-June/001398.html on how to bend a straight pipe. The code worked perfectly for bending a straight pipe into a 90 degree bend shape. The problem is I need to bend my pipe at certain section, say I have a pipe with total length of 10 and I want to bend it in the second half, i.e. from 5-10. However, when I tried to do it in this way, the error of 'Vanishing Jacobian near Xth node of element XX' appeared and the simulation couldn't start. I can't see why it doesn't work. Below is the code I'm using to bend the pipe (from 5-10) into a 90 degree bend. Hope someone could help me on this. Any suggestion would be appreciated. Thank you very much in advance. Kind regards, Tony c----------------------------------------------------------------------- subroutine usrdat2() ! This routine to modify mesh coordinates include 'SIZE' include 'TOTAL' n = nx1*ny1*nz1*nelv c First, rotate x into axial position do i=1,n x_original = xm1(i,1,1,1) y_original = ym1(i,1,1,1) z_original = zm1(i,1,1,1) xm1(i,1,1,1) = z_original ym1(i,1,1,1) = y_original zm1(i,1,1,1) = -x_original enddo c Second, bend pipe into 90 degree bend xmin = glmin(xm1,n) xmax = glmax(xm1,n) ymin = glmin(ym1,n) ymax = glmax(ym1,n) if (nid==0) print *,xmin,xmax,ymin,ymax !which are 0, 10, -1, 1 rad0 = 0.5*(ymax-ymin) ! Radius of initial pipe rad1 = 1.0 ! Radius of new pipe radm = 2.0 ! Major radius of torus do i=1,n x = xm1(i,1,1,1) y = ym1(i,1,1,1) if (x .gt. 5.0e0) then theta = 0.5*pi*(x-5.0e0)/(xmax-5.0e0) rad = radm + 2.0e0*rad1*(y-ymin)/(ymax-ymin) ym1(i,1,1,1) = rad*cos(theta) xm1(i,1,1,1) = rad*sin(theta)+5.0e0 end if end do return end -------------- 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 71, Issue 11 ********************************************* From nek5000-users at lists.mcs.anl.gov Fri Jan 23 05:17:13 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 23 Jan 2015 11:17:13 +0000 Subject: [Nek5000-users] Nek5000-users Digest, Vol 71, Issue 11 In-Reply-To: References: , Message-ID: Hi Tony, Yes -- I would just do this in usrdat2, along the lines that you currently follow. 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: Thursday, January 22, 2015 8:37 PM To: nek5000-users at lists.mcs.anl.gov Subject: Re: [Nek5000-users] Nek5000-users Digest, Vol 71, Issue 11 Dear Paul, Thank you very much for your help. Yes, the problem was the jump in y direction. I wasn't aware of the shift in y direction after bending the pipe. The code works fine now. Just a quick question, if I want to do a U-bend pipe simulation, should I just change my geometry in usrdat2 subroutine or are there any other ways you would suggest? Thanks again. Cheers, Tony ________________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov on behalf of nek5000-users-request at lists.mcs.anl.gov Sent: 22 January 2015 18:00 To: nek5000-users at lists.mcs.anl.gov Subject: Nek5000-users Digest, Vol 71, Issue 11 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. Mesh-morphing for bend pipe (nek5000-users at lists.mcs.anl.gov) 2. Re: Mesh-morphing for bend pipe (nek5000-users at lists.mcs.anl.gov) ---------------------------------------------------------------------- Message: 1 Date: Wed, 21 Jan 2015 21:53:36 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: [Nek5000-users] Mesh-morphing for bend pipe Message-ID: Content-Type: text/plain; charset="iso-8859-1" Dear Neks, I have been looking at the mesh-morphing for a bend pipe recently. I found the thread https://lists.mcs.anl.gov/mailman/htdig/nek5000-users/2011-June/001398.html on how to bend a straight pipe. The code worked perfectly for bending a straight pipe into a 90 degree bend shape. The problem is I need to bend my pipe at certain section, say I have a pipe with total length of 10 and I want to bend it in the second half, i.e. from 5-10. However, when I tried to do it in this way, the error of 'Vanishing Jacobian near Xth node of element XX' appeared and the simulation couldn't start. I can't see why it doesn't work. Below is the code I'm using to bend the pipe (from 5-10) into a 90 degree bend. Hope someone could help me on this. Any suggestion would be appreciated. Thank you very much in advance. Kind regards, Tony c----------------------------------------------------------------------- subroutine usrdat2() ! This routine to modify mesh coordinates include 'SIZE' include 'TOTAL' n = nx1*ny1*nz1*nelv c First, rotate x into axial position do i=1,n x_original = xm1(i,1,1,1) y_original = ym1(i,1,1,1) z_original = zm1(i,1,1,1) xm1(i,1,1,1) = z_original ym1(i,1,1,1) = y_original zm1(i,1,1,1) = -x_original enddo c Second, bend pipe into 90 degree bend xmin = glmin(xm1,n) xmax = glmax(xm1,n) ymin = glmin(ym1,n) ymax = glmax(ym1,n) if (nid==0) print *,xmin,xmax,ymin,ymax !which are 0, 10, -1, 1 rad0 = 0.5*(ymax-ymin) ! Radius of initial pipe rad1 = 1.0 ! Radius of new pipe radm = 2.0 ! Major radius of torus do i=1,n x = xm1(i,1,1,1) y = ym1(i,1,1,1) if (x .gt. 5.0e0) then theta = 0.5*pi*(x-5.0e0)/(xmax-5.0e0) rad = radm + 2.0e0*rad1*(y-ymin)/(ymax-ymin) ym1(i,1,1,1) = rad*cos(theta) xm1(i,1,1,1) = rad*sin(theta)+5.0e0 end if end do return end -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Message: 2 Date: Wed, 21 Jan 2015 22:51:23 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: Re: [Nek5000-users] Mesh-morphing for bend pipe Message-ID: Content-Type: text/plain; charset="iso-8859-1" Hi Tony, It appears that you shift y, but only for x > 5, so there is a sudden jump in the y value at x=5. Another possible thing that I didn't check is to make certain that your Rcos theta and R sin theta transformation preserves the right-hand-rule. (Your first rotation to x=axial direction does so.) Often the fastest way to identify these issues is to run a tiny case on one processor and just dump out all points (x,y,z) and plot them in matlab or via splot in gnuplot. hth, 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, January 21, 2015 3:53 PM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Mesh-morphing for bend pipe Dear Neks, I have been looking at the mesh-morphing for a bend pipe recently. I found the thread https://lists.mcs.anl.gov/mailman/htdig/nek5000-users/2011-June/001398.html on how to bend a straight pipe. The code worked perfectly for bending a straight pipe into a 90 degree bend shape. The problem is I need to bend my pipe at certain section, say I have a pipe with total length of 10 and I want to bend it in the second half, i.e. from 5-10. However, when I tried to do it in this way, the error of 'Vanishing Jacobian near Xth node of element XX' appeared and the simulation couldn't start. I can't see why it doesn't work. Below is the code I'm using to bend the pipe (from 5-10) into a 90 degree bend. Hope someone could help me on this. Any suggestion would be appreciated. Thank you very much in advance. Kind regards, Tony c----------------------------------------------------------------------- subroutine usrdat2() ! This routine to modify mesh coordinates include 'SIZE' include 'TOTAL' n = nx1*ny1*nz1*nelv c First, rotate x into axial position do i=1,n x_original = xm1(i,1,1,1) y_original = ym1(i,1,1,1) z_original = zm1(i,1,1,1) xm1(i,1,1,1) = z_original ym1(i,1,1,1) = y_original zm1(i,1,1,1) = -x_original enddo c Second, bend pipe into 90 degree bend xmin = glmin(xm1,n) xmax = glmax(xm1,n) ymin = glmin(ym1,n) ymax = glmax(ym1,n) if (nid==0) print *,xmin,xmax,ymin,ymax !which are 0, 10, -1, 1 rad0 = 0.5*(ymax-ymin) ! Radius of initial pipe rad1 = 1.0 ! Radius of new pipe radm = 2.0 ! Major radius of torus do i=1,n x = xm1(i,1,1,1) y = ym1(i,1,1,1) if (x .gt. 5.0e0) then theta = 0.5*pi*(x-5.0e0)/(xmax-5.0e0) rad = radm + 2.0e0*rad1*(y-ymin)/(ymax-ymin) ym1(i,1,1,1) = rad*cos(theta) xm1(i,1,1,1) = rad*sin(theta)+5.0e0 end if end do return end -------------- 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 71, Issue 11 ********************************************* _______________________________________________ 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 Fri Jan 23 12:48:16 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 23 Jan 2015 18:48:16 +0000 Subject: [Nek5000-users] Mesh-morphing for bend pipe In-Reply-To: References: Message-ID: Hi Paul, I see. Many thanks for your help. It all works fine now. Best regards, Tony ________________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov on behalf of nek5000-users-request at lists.mcs.anl.gov Sent: 23 January 2015 18:00 To: nek5000-users at lists.mcs.anl.gov Subject: Nek5000-users Digest, Vol 71, Issue 12 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: Nek5000-users Digest, Vol 71, Issue 11 (nek5000-users at lists.mcs.anl.gov) 2. Re: Nek5000-users Digest, Vol 71, Issue 11 (nek5000-users at lists.mcs.anl.gov) ---------------------------------------------------------------------- Message: 1 Date: Fri, 23 Jan 2015 02:37:53 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: Re: [Nek5000-users] Nek5000-users Digest, Vol 71, Issue 11 Message-ID: Content-Type: text/plain; charset="iso-8859-1" Dear Paul, Thank you very much for your help. Yes, the problem was the jump in y direction. I wasn't aware of the shift in y direction after bending the pipe. The code works fine now. Just a quick question, if I want to do a U-bend pipe simulation, should I just change my geometry in usrdat2 subroutine or are there any other ways you would suggest? Thanks again. Cheers, Tony ________________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov on behalf of nek5000-users-request at lists.mcs.anl.gov Sent: 22 January 2015 18:00 To: nek5000-users at lists.mcs.anl.gov Subject: Nek5000-users Digest, Vol 71, Issue 11 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. Mesh-morphing for bend pipe (nek5000-users at lists.mcs.anl.gov) 2. Re: Mesh-morphing for bend pipe (nek5000-users at lists.mcs.anl.gov) ---------------------------------------------------------------------- Message: 1 Date: Wed, 21 Jan 2015 21:53:36 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: [Nek5000-users] Mesh-morphing for bend pipe Message-ID: Content-Type: text/plain; charset="iso-8859-1" Dear Neks, I have been looking at the mesh-morphing for a bend pipe recently. I found the thread https://lists.mcs.anl.gov/mailman/htdig/nek5000-users/2011-June/001398.html on how to bend a straight pipe. The code worked perfectly for bending a straight pipe into a 90 degree bend shape. The problem is I need to bend my pipe at certain section, say I have a pipe with total length of 10 and I want to bend it in the second half, i.e. from 5-10. However, when I tried to do it in this way, the error of 'Vanishing Jacobian near Xth node of element XX' appeared and the simulation couldn't start. I can't see why it doesn't work. Below is the code I'm using to bend the pipe (from 5-10) into a 90 degree bend. Hope someone could help me on this. Any suggestion would be appreciated. Thank you very much in advance. Kind regards, Tony c----------------------------------------------------------------------- subroutine usrdat2() ! This routine to modify mesh coordinates include 'SIZE' include 'TOTAL' n = nx1*ny1*nz1*nelv c First, rotate x into axial position do i=1,n x_original = xm1(i,1,1,1) y_original = ym1(i,1,1,1) z_original = zm1(i,1,1,1) xm1(i,1,1,1) = z_original ym1(i,1,1,1) = y_original zm1(i,1,1,1) = -x_original enddo c Second, bend pipe into 90 degree bend xmin = glmin(xm1,n) xmax = glmax(xm1,n) ymin = glmin(ym1,n) ymax = glmax(ym1,n) if (nid==0) print *,xmin,xmax,ymin,ymax !which are 0, 10, -1, 1 rad0 = 0.5*(ymax-ymin) ! Radius of initial pipe rad1 = 1.0 ! Radius of new pipe radm = 2.0 ! Major radius of torus do i=1,n x = xm1(i,1,1,1) y = ym1(i,1,1,1) if (x .gt. 5.0e0) then theta = 0.5*pi*(x-5.0e0)/(xmax-5.0e0) rad = radm + 2.0e0*rad1*(y-ymin)/(ymax-ymin) ym1(i,1,1,1) = rad*cos(theta) xm1(i,1,1,1) = rad*sin(theta)+5.0e0 end if end do return end -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Message: 2 Date: Wed, 21 Jan 2015 22:51:23 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: Re: [Nek5000-users] Mesh-morphing for bend pipe Message-ID: Content-Type: text/plain; charset="iso-8859-1" Hi Tony, It appears that you shift y, but only for x > 5, so there is a sudden jump in the y value at x=5. Another possible thing that I didn't check is to make certain that your Rcos theta and R sin theta transformation preserves the right-hand-rule. (Your first rotation to x=axial direction does so.) Often the fastest way to identify these issues is to run a tiny case on one processor and just dump out all points (x,y,z) and plot them in matlab or via splot in gnuplot. hth, 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, January 21, 2015 3:53 PM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Mesh-morphing for bend pipe Dear Neks, I have been looking at the mesh-morphing for a bend pipe recently. I found the thread https://lists.mcs.anl.gov/mailman/htdig/nek5000-users/2011-June/001398.html on how to bend a straight pipe. The code worked perfectly for bending a straight pipe into a 90 degree bend shape. The problem is I need to bend my pipe at certain section, say I have a pipe with total length of 10 and I want to bend it in the second half, i.e. from 5-10. However, when I tried to do it in this way, the error of 'Vanishing Jacobian near Xth node of element XX' appeared and the simulation couldn't start. I can't see why it doesn't work. Below is the code I'm using to bend the pipe (from 5-10) into a 90 degree bend. Hope someone could help me on this. Any suggestion would be appreciated. Thank you very much in advance. Kind regards, Tony c----------------------------------------------------------------------- subroutine usrdat2() ! This routine to modify mesh coordinates include 'SIZE' include 'TOTAL' n = nx1*ny1*nz1*nelv c First, rotate x into axial position do i=1,n x_original = xm1(i,1,1,1) y_original = ym1(i,1,1,1) z_original = zm1(i,1,1,1) xm1(i,1,1,1) = z_original ym1(i,1,1,1) = y_original zm1(i,1,1,1) = -x_original enddo c Second, bend pipe into 90 degree bend xmin = glmin(xm1,n) xmax = glmax(xm1,n) ymin = glmin(ym1,n) ymax = glmax(ym1,n) if (nid==0) print *,xmin,xmax,ymin,ymax !which are 0, 10, -1, 1 rad0 = 0.5*(ymax-ymin) ! Radius of initial pipe rad1 = 1.0 ! Radius of new pipe radm = 2.0 ! Major radius of torus do i=1,n x = xm1(i,1,1,1) y = ym1(i,1,1,1) if (x .gt. 5.0e0) then theta = 0.5*pi*(x-5.0e0)/(xmax-5.0e0) rad = radm + 2.0e0*rad1*(y-ymin)/(ymax-ymin) ym1(i,1,1,1) = rad*cos(theta) xm1(i,1,1,1) = rad*sin(theta)+5.0e0 end if end do return end -------------- 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 71, Issue 11 ********************************************* ------------------------------ Message: 2 Date: Fri, 23 Jan 2015 11:17:13 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: Re: [Nek5000-users] Nek5000-users Digest, Vol 71, Issue 11 Message-ID: Content-Type: text/plain; charset="us-ascii" Hi Tony, Yes -- I would just do this in usrdat2, along the lines that you currently follow. 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: Thursday, January 22, 2015 8:37 PM To: nek5000-users at lists.mcs.anl.gov Subject: Re: [Nek5000-users] Nek5000-users Digest, Vol 71, Issue 11 Dear Paul, Thank you very much for your help. Yes, the problem was the jump in y direction. I wasn't aware of the shift in y direction after bending the pipe. The code works fine now. Just a quick question, if I want to do a U-bend pipe simulation, should I just change my geometry in usrdat2 subroutine or are there any other ways you would suggest? Thanks again. Cheers, Tony ________________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov on behalf of nek5000-users-request at lists.mcs.anl.gov Sent: 22 January 2015 18:00 To: nek5000-users at lists.mcs.anl.gov Subject: Nek5000-users Digest, Vol 71, Issue 11 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. Mesh-morphing for bend pipe (nek5000-users at lists.mcs.anl.gov) 2. Re: Mesh-morphing for bend pipe (nek5000-users at lists.mcs.anl.gov) ---------------------------------------------------------------------- Message: 1 Date: Wed, 21 Jan 2015 21:53:36 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: [Nek5000-users] Mesh-morphing for bend pipe Message-ID: Content-Type: text/plain; charset="iso-8859-1" Dear Neks, I have been looking at the mesh-morphing for a bend pipe recently. I found the thread https://lists.mcs.anl.gov/mailman/htdig/nek5000-users/2011-June/001398.html on how to bend a straight pipe. The code worked perfectly for bending a straight pipe into a 90 degree bend shape. The problem is I need to bend my pipe at certain section, say I have a pipe with total length of 10 and I want to bend it in the second half, i.e. from 5-10. However, when I tried to do it in this way, the error of 'Vanishing Jacobian near Xth node of element XX' appeared and the simulation couldn't start. I can't see why it doesn't work. Below is the code I'm using to bend the pipe (from 5-10) into a 90 degree bend. Hope someone could help me on this. Any suggestion would be appreciated. Thank you very much in advance. Kind regards, Tony c----------------------------------------------------------------------- subroutine usrdat2() ! This routine to modify mesh coordinates include 'SIZE' include 'TOTAL' n = nx1*ny1*nz1*nelv c First, rotate x into axial position do i=1,n x_original = xm1(i,1,1,1) y_original = ym1(i,1,1,1) z_original = zm1(i,1,1,1) xm1(i,1,1,1) = z_original ym1(i,1,1,1) = y_original zm1(i,1,1,1) = -x_original enddo c Second, bend pipe into 90 degree bend xmin = glmin(xm1,n) xmax = glmax(xm1,n) ymin = glmin(ym1,n) ymax = glmax(ym1,n) if (nid==0) print *,xmin,xmax,ymin,ymax !which are 0, 10, -1, 1 rad0 = 0.5*(ymax-ymin) ! Radius of initial pipe rad1 = 1.0 ! Radius of new pipe radm = 2.0 ! Major radius of torus do i=1,n x = xm1(i,1,1,1) y = ym1(i,1,1,1) if (x .gt. 5.0e0) then theta = 0.5*pi*(x-5.0e0)/(xmax-5.0e0) rad = radm + 2.0e0*rad1*(y-ymin)/(ymax-ymin) ym1(i,1,1,1) = rad*cos(theta) xm1(i,1,1,1) = rad*sin(theta)+5.0e0 end if end do return end -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Message: 2 Date: Wed, 21 Jan 2015 22:51:23 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: Re: [Nek5000-users] Mesh-morphing for bend pipe Message-ID: Content-Type: text/plain; charset="iso-8859-1" Hi Tony, It appears that you shift y, but only for x > 5, so there is a sudden jump in the y value at x=5. Another possible thing that I didn't check is to make certain that your Rcos theta and R sin theta transformation preserves the right-hand-rule. (Your first rotation to x=axial direction does so.) Often the fastest way to identify these issues is to run a tiny case on one processor and just dump out all points (x,y,z) and plot them in matlab or via splot in gnuplot. hth, 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, January 21, 2015 3:53 PM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Mesh-morphing for bend pipe Dear Neks, I have been looking at the mesh-morphing for a bend pipe recently. I found the thread https://lists.mcs.anl.gov/mailman/htdig/nek5000-users/2011-June/001398.html on how to bend a straight pipe. The code worked perfectly for bending a straight pipe into a 90 degree bend shape. The problem is I need to bend my pipe at certain section, say I have a pipe with total length of 10 and I want to bend it in the second half, i.e. from 5-10. However, when I tried to do it in this way, the error of 'Vanishing Jacobian near Xth node of element XX' appeared and the simulation couldn't start. I can't see why it doesn't work. Below is the code I'm using to bend the pipe (from 5-10) into a 90 degree bend. Hope someone could help me on this. Any suggestion would be appreciated. Thank you very much in advance. Kind regards, Tony c----------------------------------------------------------------------- subroutine usrdat2() ! This routine to modify mesh coordinates include 'SIZE' include 'TOTAL' n = nx1*ny1*nz1*nelv c First, rotate x into axial position do i=1,n x_original = xm1(i,1,1,1) y_original = ym1(i,1,1,1) z_original = zm1(i,1,1,1) xm1(i,1,1,1) = z_original ym1(i,1,1,1) = y_original zm1(i,1,1,1) = -x_original enddo c Second, bend pipe into 90 degree bend xmin = glmin(xm1,n) xmax = glmax(xm1,n) ymin = glmin(ym1,n) ymax = glmax(ym1,n) if (nid==0) print *,xmin,xmax,ymin,ymax !which are 0, 10, -1, 1 rad0 = 0.5*(ymax-ymin) ! Radius of initial pipe rad1 = 1.0 ! Radius of new pipe radm = 2.0 ! Major radius of torus do i=1,n x = xm1(i,1,1,1) y = ym1(i,1,1,1) if (x .gt. 5.0e0) then theta = 0.5*pi*(x-5.0e0)/(xmax-5.0e0) rad = radm + 2.0e0*rad1*(y-ymin)/(ymax-ymin) ym1(i,1,1,1) = rad*cos(theta) xm1(i,1,1,1) = rad*sin(theta)+5.0e0 end if end do return end -------------- 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 71, Issue 11 ********************************************* _______________________________________________ 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 71, Issue 12 ********************************************* From nek5000-users at lists.mcs.anl.gov Sat Jan 24 12:08:27 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sat, 24 Jan 2015 19:08:27 +0100 Subject: [Nek5000-users] uniform interpolation Message-ID: For time-space Fourier analysis I need data on a uniform mesh. So I have used int_tp. When I applied int_tp to binary *0.f* files (param(66) = 6) I have got a strange pictures which reflect the solution, but as if some elements were switched before interpolation. So I tried to make uniform interpolation to the most simple tutorial examples, namely Rayleigh-Bendard ray_nn case. Again if I apply int_tp to binary *0.f* of *fld*, then the result is as if the some space patches where switched, but if int_tp is applied ot ASCII fld files (param(66) = 6) then int_tp works and saves the result correctly to both text and binary format. Is it a common problem, or I miss something? (input files in text format work for me but occupie too much space.) Also, for each file in file.list int_tp asks for input file format. But to make time-space analysis I need more that 1000 files in file.list, and it is tiresome to type an answer that number of times. It could be better to choose input file format just once at the beginning? Thank you in advance, Ilias -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Sat Jan 24 20:05:13 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sat, 24 Jan 2015 19:05:13 -0700 Subject: [Nek5000-users] uniform interpolation In-Reply-To: References: Message-ID: Hi Ilias, For the file.list, you could modify the script below, in order to circumvent, writing too mainy files in the "file.list" *--------------------------------------------------------------------------------* program lst1 implicit none integer, parameter :: N = 140 character*80,dimension(N) :: filename integer :: i,j,k open(unit = 10,file='file.list',action='write') write(10,"(I3)") N do i = 1,N if(i .lt. 100)then write(filename(i),'(A,I5.5)') 'turbChannel0.f',i else write(filename(i),'(A,I5.5)') 'turbChannel0.f',i endif enddo do i = 1,N write(10,*) trim(filename(i)) enddo close(10) stop end program lst1 *--------------------------------------------------------------------------------* For various post-processing format I use this script to write multiple files in file-list. I was previously using ifreguo = .true. in post-processing format of nek for uniform interpolation. But, how are you performing FFT after interpolating to uniform mesh. Are you using a library, could you elucidate/ describe a little more on the process, on how you do that? I was trying to do that as well. Sincerely, Tanmoy Best Regards, Tanmoy On Sat, Jan 24, 2015 at 11:08 AM, wrote: > > For time-space Fourier analysis I need data on a uniform mesh. > So I have used int_tp. > > When I applied int_tp to binary *0.f* files (param(66) = 6) I have got a > strange pictures which reflect the solution, but as if some elements were > switched before interpolation. > > So I tried to make uniform interpolation to the most simple tutorial > examples, namely Rayleigh-Bendard ray_nn case. > > Again if I apply int_tp to binary *0.f* of *fld*, then the result is as if > the some space patches where switched, > but if int_tp is applied ot ASCII fld files (param(66) = 6) then int_tp > works and saves the result correctly to both text and binary format. > > Is it a common problem, or I miss something? (input files in text format > work for me but occupie too much space.) > > Also, for each file in file.list int_tp asks for input file format. > But to make time-space analysis I need more that 1000 files in file.list, > and > it is tiresome to type an answer that number of times. > It could be better to choose input file format just once at the beginning? > > Thank you in advance, 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 Sat Jan 24 22:07:03 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sat, 24 Jan 2015 23:07:03 -0500 Subject: [Nek5000-users] nek5000 compilation error of on NERSC edison/hopper Message-ID: Dear NEK users, I am trying to compile nek model on NERSC machines edison/hopper, encountering an error. Tail of the compilation message is as below: _____________________________________________________________________ /global/homes/r/rtumkur/nek5_svn/trunk/nek/prepost.f(2482): warning #5117: Bad # preprocessor line #ifdef MPIIO -^ /global/homes/r/rtumkur/nek5_svn/trunk/nek/prepost.f(2484): warning #5117: Bad # preprocessor line #else -^ /global/homes/r/rtumkur/nek5_svn/trunk/nek/prepost.f(2486): warning #5117: Bad # preprocessor line #endif -^ compilation aborted for /global/homes/r/rtumkur/nek5_svn/trunk/nek/prepost.f (code 1) make: *** [obj/prepost.o] Error 1 make: *** Waiting for unfinished jobs.... ifort: command line warning #10006: ignoring unknown option '-Mpreprocess' ____________________________________________________________________ Here are the steps I have taken leading up the error. I downloaded the 1. Downloaded and build the nek svn co https://svn.mcs.anl.gov/repos/nek5/ nek5_svn cd nek5_svn/trunk/tools/ ./maketools all 2. Updated the makenek file as per the instructions provide at the link https://www.nersc.gov/users/computational-systems/hopper/programming/compiling-codes-on-hopper/#toc-anchor-6 Accordingly, I have loaded Intel compiler module and specified the wrapper in makenek: # Fortran compiler F77="ftn" # C compiler CC="cc" Also included OpenMP Compilation in the general compiler flag: # generic compiler flags G="-g -openmp" 3. Compiled the code makenek sprcyl Any help in resolving this error is much appreciated. I have inserted the SIZE file and makenek file for your reference. ____________________SIZE___________________________ C Dimension file to be included C C HCUBE array dimensions C parameter (ldim=2) parameter (lx1=6,ly1=lx1,lz1=1,lelt=1500,lelv=lelt) parameter (lxd=9,lyd=lxd,lzd=1) parameter (lelx=1,lely=1,lelz=1) parameter (lp = 32768) parameter (lelg = 50000) parameter (lzl=3 + 2*(ldim-3)) c parameter (lx2=lx1-2) parameter (ly2=ly1-2) parameter (lz2=lz1 ) parameter (lx3=lx1) parameter (ly3=ly1) parameter (lz3=lz1) c c parameter (lpelv=lelv,lpelt=lelt,lpert=3) ! perturbation c parameter (lpx1=lx1,lpy1=ly1,lpz1=lz1) ! array sizes c parameter (lpx2=lx2,lpy2=ly2,lpz2=lz2) c parameter (lpelv=1,lpelt=1,lpert=1) ! perturbation parameter (lpx1=1,lpy1=1,lpz1=1) ! array sizes parameter (lpx2=1,lpy2=1,lpz2=1) c c c parameter (lbelv=lelv,lbelt=lelt) ! MHD c parameter (lbx1=lx1,lby1=ly1,lbz1=lz1) ! array sizes c parameter (lbx2=lx2,lby2=ly2,lbz2=lz2) c parameter (lbelv=1,lbelt=1) ! MHD parameter (lbx1=1,lby1=1,lbz1=1) ! array sizes parameter (lbx2=1,lby2=1,lbz2=1) c C LX1M=LX1 when there are moving meshes; =1 otherwise parameter (lx1m=lx1,ly1m=ly1,lz1m=lz1) parameter (ldimt= 1) ! 3 passive scalars + T parameter (ldimt1=ldimt+1) parameter (ldimt3=ldimt+3) c c Note: In the new code, LELGEC should be about sqrt(LELG) c PARAMETER (LELGEC = 1) PARAMETER (LXYZ2 = 1) PARAMETER (LXZ21 = 1) c PARAMETER (LMAXV=LX1*LY1*LZ1*LELV) PARAMETER (LMAXT=LX1*LY1*LZ1*LELT) PARAMETER (LMAXP=LX2*LY2*LZ2*LELV) PARAMETER (LXZ=LX1*LZ1) PARAMETER (LORDER=3) PARAMETER (MAXOBJ=4,MAXMBR=LELT*6,lhis=100) C C Common Block Dimensions C PARAMETER (LCTMP0 =2*LX1*LY1*LZ1*LELT) PARAMETER (LCTMP1 =4*LX1*LY1*LZ1*LELT) C C The parameter LVEC controls whether an additional 42 field arrays C are required for Steady State Solutions. If you are not using C Steady State, it is recommended that LVEC=1. C PARAMETER (LVEC=1) C C Uzawa projection array dimensions C parameter (mxprev = 20) parameter (lgmres = 20) C C Split projection array dimensions C parameter(lmvec = 1) parameter(lsvec = 1) parameter(lstore=lmvec*lsvec) c c NONCONFORMING STUFF c parameter (maxmor = lelt) C C Array dimensions C COMMON/DIMN/NELV,NELT,NX1,NY1,NZ1,NX2,NY2,NZ2 $,NX3,NY3,NZ3,NDIM,NFIELD,NPERT,NID $,NXD,NYD,NZD c automatically added by makenek parameter(lxo = lx1) ! max output grid size (lxo>=lx1) c automatically added by makenek parameter(lpart = 1 ) ! max number of particles c automatically added by makenek integer ax1,ay1,az1,ax2,ay2,az2 parameter (ax1=lx1,ay1=ly1,az1=lz1,ax2=lx2,ay2=ly2,az2=lz2) ! running averages c automatically added by makenek parameter (lxs=1,lys=lxs,lzs=(lxs-1)*(ldim-2)+1) !New Pressure Preconditioner c automatically added by makenek parameter (lfdm=0) ! == 1 for fast diagonalization method c automatically added by makenek common/IOFLAG/nio ! for logfile verbosity control _____________________makenek (initial part)__________________ !/bin/bash # Nek5000 build config file # (c) 2008,2009,2010 UCHICAGO ARGONNE, LLC # source path SOURCE_ROOT="$HOME/nek5_svn/trunk/nek" # Fortran compiler F77="ftn" # C compiler CC="cc" # pre-processor symbol list # (set PPLIST=? to get a list of available symbols) #PPLIST="-openmp" # plug-in list PLUGIN_LIST="" # OPTIONAL SETTINGS # ----------------- # enable MPI (default true) #IFMPI="false" # auxilliary files to compile # NOTE: source files have to located in the same directory as makenek # a makefile_usr.inc has to be provided containing the build rules #USR="foo.o" # linking flags #USR_LFLAGS="-L/usr/lib -lfoo" # generic compiler flags G="-g -openmp" # optimization flags #OPT_FLAGS_STD="" #OPT_FLAGS_MAG="" # enable AMG coarse grid solver (default XXT) #IFAMG="true" #IFAMG_DUMP="true" # CVODE path #CVODE_DIR=$HOME/cvode/lib # MOAB/iMESH path #MOAB_DIR="$HOME/moab" # For linking to MOAB, the following might be needed: # NOTE: compiler specific, use the appropriate one depending on your compiler #GNU="-fopenmp" -- Ravi Kumar R. Tumkur, PhD. Phone: 217 778 7538 -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Mon Jan 26 08:28:43 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 26 Jan 2015 15:28:43 +0100 Subject: [Nek5000-users] uniform interpolation Message-ID: Hi Tanmoy, thanks for the answer int_tp is working perfectly for me now, the only issue is that it works only for ASCII input files. For my 3D simulation each text file occupies 100mb and I need more than 1000 time slices for analysis, so it becomes a real issue. 1. I have not understood the purpose of you script, to make file.list I just execute a command like ls -1 *0.f* > file.list What was annoying for me is the manual confirmations after each file in list. I got rid of that by moving the part of int_tp.f file: --- write(6,*) 'Input File Format: 0 - ASCII, 1 - binary(fld), --- $ 2 - binary(0.f0000?) ? ' --- read (5,*) ib --- ifbyte=.false. --- iffbin=.false. --- if(ib.ne.0) ifbyte = .true. --- if(ib.eq.2) iffbin = .true. out of the cycle. But what I can't figure out is how to make it working properly with binary input format 2. If I am correct ifreguo = .true. allows you write on a uniform mesh inside each element instead of GLL points. But I have a nonuniform mesh in terms of elements -- the density of elements is greater near the boundaries. So uniform mesh inside elements will not give me global uniform mesh. 3. For Fourier analysis I use either fftw (www.fftw.org) or routines from matlab, which are based on fftw. Really i need to filter spacial waves on different time frequency, and for that I need both time and space analysis (Hilbert transform), which can be also done with direct and inverse fft. >From results of int_tp I can construct a vector or square matrix (in 2D case) in matlab and apply fft to it. Best regards, Ilias Hi Ilias to circumvent, writing too mainy files in the "file.list" > > > *--------------------------------------------------------------------------------* > program lst1 > implicit none > > integer, parameter :: N = 140 > character*80,dimension(N) :: filename > integer :: i,j,k > > > open(unit = 10,file='file.list',action='write') > write(10,"(I3)") N > do i = 1,N > > if(i .lt. 100)then > write(filename(i),'(A,I5.5)') 'turbChannel0.f',i > else > write(filename(i),'(A,I5.5)') 'turbChannel0.f',i > endif > enddo > > do i = 1,N > write(10,*) trim(filename(i)) > enddo > > > close(10) > stop > end program lst1 > > *--------------------------------------------------------------------------------* > > For various post-processing format I use this script to write multiple > files in file-list. > > > I was previously using ifreguo = .true. in post-processing format of nek > for uniform interpolation. But, how are you performing FFT after > interpolating to uniform mesh. Are you using a library, could you > elucidate/ describe a little more on the process, on how you do that? I was > trying to do that as well. > > Sincerely, > Tanmoy > > Best Regards, > Tanmoy > > On Sat, Jan 24, 2015 at 11:08 AM, wrote: > >> >> For time-space Fourier analysis I need data on a uniform mesh. >> So I have used int_tp. >> >> When I applied int_tp to binary *0.f* files (param(66) = 6) I have got a >> strange pictures which reflect the solution, but as if some elements were >> switched before interpolation. >> >> So I tried to make uniform interpolation to the most simple tutorial >> examples, namely Rayleigh-Bendard ray_nn case. >> >> Again if I apply int_tp to binary *0.f* of *fld*, then the result is as >> if the some space patches where switched, >> but if int_tp is applied ot ASCII fld files (param(66) = 6) then int_tp >> works and saves the result correctly to both text and binary format. >> >> Is it a common problem, or I miss something? (input files in text format >> work for me but occupie too much space.) >> >> Also, for each file in file.list int_tp asks for input file format. >> But to make time-space analysis I need more that 1000 files in file.list, >> and >> it is tiresome to type an answer that number of times. >> It could be better to choose input file format just once at the beginning? >> >> Thank you in advance, 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 Mon Jan 26 10:48:32 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Mon, 26 Jan 2015 09:48:32 -0700 Subject: [Nek5000-users] uniform interpolation In-Reply-To: References: Message-ID: Hi Illias, Thanks for the answer and clarification. Sincerely, Tanmoy Best Regards, Tanmoy On Mon, Jan 26, 2015 at 7:28 AM, wrote: > Hi Tanmoy, thanks for the answer > > > int_tp is working perfectly for me now, the only issue is that it works > only for ASCII input files. > > For my 3D simulation each text file occupies 100mb and I need more than > 1000 time slices for analysis, > > so it becomes a real issue. > > > 1. I have not understood the purpose of you script, > > to make file.list I just execute a command like > > ls -1 *0.f* > file.list > > What was annoying for me is the manual confirmations after each file in > list. > > I got rid of that by moving the part of int_tp.f file: > > --- write(6,*) 'Input File Format: 0 - ASCII, 1 - binary(fld), > > --- $ 2 - binary(0.f0000?) ? ' > > --- read (5,*) ib > > --- ifbyte=.false. > > --- iffbin=.false. > > --- if(ib.ne.0) ifbyte = .true. > > --- if(ib.eq.2) iffbin = .true. > > out of the cycle. > > But what I can't figure out is how to make it working properly with > binary input format > > > > 2. If I am correct ifreguo = .true. allows you write on a uniform mesh > inside each element > > instead of GLL points. But I have a nonuniform mesh in terms of elements -- > > the density of elements is greater near the boundaries. > > So uniform mesh inside elements will not give me global uniform mesh. > > > 3. For Fourier analysis I use either fftw (www.fftw.org) > > or routines from matlab, which are based on fftw. > > Really i need to filter spacial waves on different time frequency, and for > that I need both time and space analysis (Hilbert transform), > > which can be also done with direct and inverse fft. > > From results of int_tp I can construct a vector or square matrix (in 2D > case) in matlab and apply fft to it. > > > Best regards, Ilias > > > > > > Hi Ilias to circumvent, writing too mainy files in the "file.list" >> >> >> *--------------------------------------------------------------------------------* >> program lst1 >> implicit none >> >> integer, parameter :: N = 140 >> character*80,dimension(N) :: filename >> integer :: i,j,k >> >> >> open(unit = 10,file='file.list',action='write') >> write(10,"(I3)") N >> do i = 1,N >> >> if(i .lt. 100)then >> write(filename(i),'(A,I5.5)') 'turbChannel0.f',i >> else >> write(filename(i),'(A,I5.5)') 'turbChannel0.f',i >> endif >> enddo >> >> do i = 1,N >> write(10,*) trim(filename(i)) >> enddo >> >> >> close(10) >> stop >> end program lst1 >> >> *--------------------------------------------------------------------------------* >> >> For various post-processing format I use this script to write multiple >> files in file-list. >> >> >> I was previously using ifreguo = .true. in post-processing format of nek >> for uniform interpolation. But, how are you performing FFT after >> interpolating to uniform mesh. Are you using a library, could you >> elucidate/ describe a little more on the process, on how you do that? I was >> trying to do that as well. >> >> Sincerely, >> Tanmoy >> >> Best Regards, >> Tanmoy >> >> On Sat, Jan 24, 2015 at 11:08 AM, >> wrote: >> >>> >>> For time-space Fourier analysis I need data on a uniform mesh. >>> So I have used int_tp. >>> >>> When I applied int_tp to binary *0.f* files (param(66) = 6) I have got >>> a strange pictures which reflect the solution, but as if some elements were >>> switched before interpolation. >>> >>> So I tried to make uniform interpolation to the most simple tutorial >>> examples, namely Rayleigh-Bendard ray_nn case. >>> >>> Again if I apply int_tp to binary *0.f* of *fld*, then the result is as >>> if the some space patches where switched, >>> but if int_tp is applied ot ASCII fld files (param(66) = 6) then int_tp >>> works and saves the result correctly to both text and binary format. >>> >>> Is it a common problem, or I miss something? (input files in text format >>> work for me but occupie too much space.) >>> >>> Also, for each file in file.list int_tp asks for input file format. >>> But to make time-space analysis I need more that 1000 files in >>> file.list, and >>> it is tiresome to type an answer that number of times. >>> It could be better to choose input file format just once at the >>> beginning? >>> >>> Thank you in advance, 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 >> >> > > _______________________________________________ > 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 Jan 29 09:50:43 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 29 Jan 2015 15:50:43 +0000 Subject: [Nek5000-users] Error in Hmholtz solving Message-ID: Dear Neks, I'm getting the error message of '**ERROR**: Failed in HMHOLTZ: ' when running my simulation. I've checked the subroutine 'cggo' in hmholtz.f where the error message is printed. I understand that when the tolerance is not satisfied, the code would output this error message. But what I'm wondering is why the maximum iteration number is only 100. In the subroutine 'cggo', niter is equal to min(maxit,maxcg) where maxit=200 and maxcg=900. I can't see other places where maxit is restricted to only 100. So why the maximum iteration number is 100 when solving the Hmholtz equation? Also, is it right that if I increase the maximum number of iterations to certain point where the tolerance is satisfied, I won't get the error message? Hope anyone can help me on this. Thank you very much in advance. Best regards, Tony Example of the error message: Step 52, t= 2.5600000E+01, DT= 5.0000000E-02, C= 2.689 5.6747E+02 1.0237E+01 Solving for fluid T T T 52 PRES gmres: 29 9.7057E-07 1.0000E-06 1.5219E-03 1.0926E+00 2.1480E+00 F 52 Hmholtz VELX: 99 8.5954E-10 6.0898E-03 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELY 1.7088E-09 2.5049E-02 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4665E-09 2.5165E-02 1.0000E-09 L1/L2 DIV(V) : 2.7573E-18 9.1661E-04 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.7573E-18 9.1661E-04 52 2.5600E+01 1.0094E+01 Fluid done Step 53, t= 2.5650000E+01, DT= 5.0000000E-02, C= 2.697 5.7766E+02 1.0186E+01 Solving for fluid T T T 53 PRES gmres: 28 9.6076E-07 1.0000E-06 1.1322E-03 1.0558E+00 2.0542E+00 F 53 Hmholtz VELX: 98 9.4891E-10 5.8783E-03 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELY 1.6450E-09 2.4941E-02 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4160E-09 2.5049E-02 1.0000E-09 L1/L2 DIV(V) : 2.4695E-18 1.0845E-03 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.4695E-18 1.0845E-03 53 2.5650E+01 9.9794E+00 Fluid done -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu Jan 29 10:00:45 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 29 Jan 2015 16:00:45 +0000 Subject: [Nek5000-users] Error in Hmholtz solving In-Reply-To: References: Message-ID: Tony, It looks like your Helmholtz tolerance is too tight. If your problem is scaled in convective time units with characteristic length scale 1, timescale 1, and visc ~ 1/Re, then typical Helmholtz tolerances are around 1.e-7. You might also try turning on projection --- set p93=20 (and lmxprev=20 in SIZE), p94=5 and p95=5 in the rea file. No guarantees at this point because there seems to be something flaky in the projection routines these days under certain runtime combinations (e.g., I think splitting + heat transfer + velocity projection does not work)... It's on my todo list to clean these up. Note that your helmholtz solves are pretty well converged... residual is 1.7e-9, etc. Projection should easily put you over the top 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: Thursday, January 29, 2015 9:50 AM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Error in Hmholtz solving Dear Neks, I'm getting the error message of '**ERROR**: Failed in HMHOLTZ: ' when running my simulation. I've checked the subroutine 'cggo' in hmholtz.f where the error message is printed. I understand that when the tolerance is not satisfied, the code would output this error message. But what I'm wondering is why the maximum iteration number is only 100. In the subroutine 'cggo', niter is equal to min(maxit,maxcg) where maxit=200 and maxcg=900. I can't see other places where maxit is restricted to only 100. So why the maximum iteration number is 100 when solving the Hmholtz equation? Also, is it right that if I increase the maximum number of iterations to certain point where the tolerance is satisfied, I won't get the error message? Hope anyone can help me on this. Thank you very much in advance. Best regards, Tony Example of the error message: Step 52, t= 2.5600000E+01, DT= 5.0000000E-02, C= 2.689 5.6747E+02 1.0237E+01 Solving for fluid T T T 52 PRES gmres: 29 9.7057E-07 1.0000E-06 1.5219E-03 1.0926E+00 2.1480E+00 F 52 Hmholtz VELX: 99 8.5954E-10 6.0898E-03 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELY 1.7088E-09 2.5049E-02 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4665E-09 2.5165E-02 1.0000E-09 L1/L2 DIV(V) : 2.7573E-18 9.1661E-04 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.7573E-18 9.1661E-04 52 2.5600E+01 1.0094E+01 Fluid done Step 53, t= 2.5650000E+01, DT= 5.0000000E-02, C= 2.697 5.7766E+02 1.0186E+01 Solving for fluid T T T 53 PRES gmres: 28 9.6076E-07 1.0000E-06 1.1322E-03 1.0558E+00 2.0542E+00 F 53 Hmholtz VELX: 98 9.4891E-10 5.8783E-03 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELY 1.6450E-09 2.4941E-02 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4160E-09 2.5049E-02 1.0000E-09 L1/L2 DIV(V) : 2.4695E-18 1.0845E-03 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.4695E-18 1.0845E-03 53 2.5650E+01 9.9794E+00 Fluid done -------------- next part -------------- An HTML attachment was scrubbed... URL: From nek5000-users at lists.mcs.anl.gov Thu Jan 29 13:44:28 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 29 Jan 2015 19:44:28 +0000 Subject: [Nek5000-users] Error in Hmholtz solving In-Reply-To: References: Message-ID: Hi Paul, Many thanks for your help. Yes, my simulation is scaled in convective time units. As you pointed out, the solutions are well converged. I just have a question about the projection and dealiasing. I'm still not sure when I need to use them in my simulations. Say I'm doing simulations without any models, do I need to turn on projection and dealiasing? I actually don't know how they will affect the solutions, would projection or dealiasing (off, old or new?) change the solutions? It would be great if you could briefly explain this bit to me. 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: 29 January 2015 18:00 To: nek5000-users at lists.mcs.anl.gov Subject: Nek5000-users Digest, Vol 71, 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. Error in Hmholtz solving (nek5000-users at lists.mcs.anl.gov) 2. Re: Error in Hmholtz solving (nek5000-users at lists.mcs.anl.gov) ---------------------------------------------------------------------- Message: 1 Date: Thu, 29 Jan 2015 15:50:43 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: [Nek5000-users] Error in Hmholtz solving Message-ID: Content-Type: text/plain; charset="iso-8859-1" Dear Neks, I'm getting the error message of '**ERROR**: Failed in HMHOLTZ: ' when running my simulation. I've checked the subroutine 'cggo' in hmholtz.f where the error message is printed. I understand that when the tolerance is not satisfied, the code would output this error message. But what I'm wondering is why the maximum iteration number is only 100. In the subroutine 'cggo', niter is equal to min(maxit,maxcg) where maxit=200 and maxcg=900. I can't see other places where maxit is restricted to only 100. So why the maximum iteration number is 100 when solving the Hmholtz equation? Also, is it right that if I increase the maximum number of iterations to certain point where the tolerance is satisfied, I won't get the error message? Hope anyone can help me on this. Thank you very much in advance. Best regards, Tony Example of the error message: Step 52, t= 2.5600000E+01, DT= 5.0000000E-02, C= 2.689 5.6747E+02 1.0237E+01 Solving for fluid T T T 52 PRES gmres: 29 9.7057E-07 1.0000E-06 1.5219E-03 1.0926E+00 2.1480E+00 F 52 Hmholtz VELX: 99 8.5954E-10 6.0898E-03 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELY 1.7088E-09 2.5049E-02 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4665E-09 2.5165E-02 1.0000E-09 L1/L2 DIV(V) : 2.7573E-18 9.1661E-04 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.7573E-18 9.1661E-04 52 2.5600E+01 1.0094E+01 Fluid done Step 53, t= 2.5650000E+01, DT= 5.0000000E-02, C= 2.697 5.7766E+02 1.0186E+01 Solving for fluid T T T 53 PRES gmres: 28 9.6076E-07 1.0000E-06 1.1322E-03 1.0558E+00 2.0542E+00 F 53 Hmholtz VELX: 98 9.4891E-10 5.8783E-03 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELY 1.6450E-09 2.4941E-02 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4160E-09 2.5049E-02 1.0000E-09 L1/L2 DIV(V) : 2.4695E-18 1.0845E-03 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.4695E-18 1.0845E-03 53 2.5650E+01 9.9794E+00 Fluid done -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Message: 2 Date: Thu, 29 Jan 2015 16:00:45 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: Re: [Nek5000-users] Error in Hmholtz solving Message-ID: Content-Type: text/plain; charset="iso-8859-1" Tony, It looks like your Helmholtz tolerance is too tight. If your problem is scaled in convective time units with characteristic length scale 1, timescale 1, and visc ~ 1/Re, then typical Helmholtz tolerances are around 1.e-7. You might also try turning on projection --- set p93=20 (and lmxprev=20 in SIZE), p94=5 and p95=5 in the rea file. No guarantees at this point because there seems to be something flaky in the projection routines these days under certain runtime combinations (e.g., I think splitting + heat transfer + velocity projection does not work)... It's on my todo list to clean these up. Note that your helmholtz solves are pretty well converged... residual is 1.7e-9, etc. Projection should easily put you over the top 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: Thursday, January 29, 2015 9:50 AM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Error in Hmholtz solving Dear Neks, I'm getting the error message of '**ERROR**: Failed in HMHOLTZ: ' when running my simulation. I've checked the subroutine 'cggo' in hmholtz.f where the error message is printed. I understand that when the tolerance is not satisfied, the code would output this error message. But what I'm wondering is why the maximum iteration number is only 100. In the subroutine 'cggo', niter is equal to min(maxit,maxcg) where maxit=200 and maxcg=900. I can't see other places where maxit is restricted to only 100. So why the maximum iteration number is 100 when solving the Hmholtz equation? Also, is it right that if I increase the maximum number of iterations to certain point where the tolerance is satisfied, I won't get the error message? Hope anyone can help me on this. Thank you very much in advance. Best regards, Tony Example of the error message: Step 52, t= 2.5600000E+01, DT= 5.0000000E-02, C= 2.689 5.6747E+02 1.0237E+01 Solving for fluid T T T 52 PRES gmres: 29 9.7057E-07 1.0000E-06 1.5219E-03 1.0926E+00 2.1480E+00 F 52 Hmholtz VELX: 99 8.5954E-10 6.0898E-03 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELY 1.7088E-09 2.5049E-02 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4665E-09 2.5165E-02 1.0000E-09 L1/L2 DIV(V) : 2.7573E-18 9.1661E-04 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.7573E-18 9.1661E-04 52 2.5600E+01 1.0094E+01 Fluid done Step 53, t= 2.5650000E+01, DT= 5.0000000E-02, C= 2.697 5.7766E+02 1.0186E+01 Solving for fluid T T T 53 PRES gmres: 28 9.6076E-07 1.0000E-06 1.1322E-03 1.0558E+00 2.0542E+00 F 53 Hmholtz VELX: 98 9.4891E-10 5.8783E-03 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELY 1.6450E-09 2.4941E-02 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4160E-09 2.5049E-02 1.0000E-09 L1/L2 DIV(V) : 2.4695E-18 1.0845E-03 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.4695E-18 1.0845E-03 53 2.5650E+01 9.9794E+00 Fluid done -------------- 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 71, Issue 16 ********************************************* From nek5000-users at lists.mcs.anl.gov Thu Jan 29 14:05:04 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Thu, 29 Jan 2015 20:05:04 +0000 Subject: [Nek5000-users] Error in Hmholtz solving In-Reply-To: References: , Message-ID: Tony, Projection simply accelerates solution of the linear systems -- see: http://www.mcs.anl.gov/~fischer/pubhtml/project.pdf Dealiasing is invariably a good idea and not too expensive. It just ensure proper integration of the nonlinear terms and avoids some spurious eigenvalues. J. Malm, P. Schlatter, P. F. Fischer, and D. S. Henningson. Stabilization of the spectral element method in convection dominated flows by recovery of skew-symmetry. J. Sci. Comput., 57:254? 277, 2013. The real question is whether to use filtering or not... I generally do, and I do so by setting p103=0.01 and p101=0 in cases that are well-resolved. 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: Thursday, January 29, 2015 1:44 PM To: nek5000-users at lists.mcs.anl.gov Subject: Re: [Nek5000-users] Error in Hmholtz solving Hi Paul, Many thanks for your help. Yes, my simulation is scaled in convective time units. As you pointed out, the solutions are well converged. I just have a question about the projection and dealiasing. I'm still not sure when I need to use them in my simulations. Say I'm doing simulations without any models, do I need to turn on projection and dealiasing? I actually don't know how they will affect the solutions, would projection or dealiasing (off, old or new?) change the solutions? It would be great if you could briefly explain this bit to me. 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: 29 January 2015 18:00 To: nek5000-users at lists.mcs.anl.gov Subject: Nek5000-users Digest, Vol 71, 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. Error in Hmholtz solving (nek5000-users at lists.mcs.anl.gov) 2. Re: Error in Hmholtz solving (nek5000-users at lists.mcs.anl.gov) ---------------------------------------------------------------------- Message: 1 Date: Thu, 29 Jan 2015 15:50:43 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: [Nek5000-users] Error in Hmholtz solving Message-ID: Content-Type: text/plain; charset="iso-8859-1" Dear Neks, I'm getting the error message of '**ERROR**: Failed in HMHOLTZ: ' when running my simulation. I've checked the subroutine 'cggo' in hmholtz.f where the error message is printed. I understand that when the tolerance is not satisfied, the code would output this error message. But what I'm wondering is why the maximum iteration number is only 100. In the subroutine 'cggo', niter is equal to min(maxit,maxcg) where maxit=200 and maxcg=900. I can't see other places where maxit is restricted to only 100. So why the maximum iteration number is 100 when solving the Hmholtz equation? Also, is it right that if I increase the maximum number of iterations to certain point where the tolerance is satisfied, I won't get the error message? Hope anyone can help me on this. Thank you very much in advance. Best regards, Tony Example of the error message: Step 52, t= 2.5600000E+01, DT= 5.0000000E-02, C= 2.689 5.6747E+02 1.0237E+01 Solving for fluid T T T 52 PRES gmres: 29 9.7057E-07 1.0000E-06 1.5219E-03 1.0926E+00 2.1480E+00 F 52 Hmholtz VELX: 99 8.5954E-10 6.0898E-03 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELY 1.7088E-09 2.5049E-02 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4665E-09 2.5165E-02 1.0000E-09 L1/L2 DIV(V) : 2.7573E-18 9.1661E-04 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.7573E-18 9.1661E-04 52 2.5600E+01 1.0094E+01 Fluid done Step 53, t= 2.5650000E+01, DT= 5.0000000E-02, C= 2.697 5.7766E+02 1.0186E+01 Solving for fluid T T T 53 PRES gmres: 28 9.6076E-07 1.0000E-06 1.1322E-03 1.0558E+00 2.0542E+00 F 53 Hmholtz VELX: 98 9.4891E-10 5.8783E-03 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELY 1.6450E-09 2.4941E-02 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4160E-09 2.5049E-02 1.0000E-09 L1/L2 DIV(V) : 2.4695E-18 1.0845E-03 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.4695E-18 1.0845E-03 53 2.5650E+01 9.9794E+00 Fluid done -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Message: 2 Date: Thu, 29 Jan 2015 16:00:45 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: Re: [Nek5000-users] Error in Hmholtz solving Message-ID: Content-Type: text/plain; charset="iso-8859-1" Tony, It looks like your Helmholtz tolerance is too tight. If your problem is scaled in convective time units with characteristic length scale 1, timescale 1, and visc ~ 1/Re, then typical Helmholtz tolerances are around 1.e-7. You might also try turning on projection --- set p93=20 (and lmxprev=20 in SIZE), p94=5 and p95=5 in the rea file. No guarantees at this point because there seems to be something flaky in the projection routines these days under certain runtime combinations (e.g., I think splitting + heat transfer + velocity projection does not work)... It's on my todo list to clean these up. Note that your helmholtz solves are pretty well converged... residual is 1.7e-9, etc. Projection should easily put you over the top 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: Thursday, January 29, 2015 9:50 AM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Error in Hmholtz solving Dear Neks, I'm getting the error message of '**ERROR**: Failed in HMHOLTZ: ' when running my simulation. I've checked the subroutine 'cggo' in hmholtz.f where the error message is printed. I understand that when the tolerance is not satisfied, the code would output this error message. But what I'm wondering is why the maximum iteration number is only 100. In the subroutine 'cggo', niter is equal to min(maxit,maxcg) where maxit=200 and maxcg=900. I can't see other places where maxit is restricted to only 100. So why the maximum iteration number is 100 when solving the Hmholtz equation? Also, is it right that if I increase the maximum number of iterations to certain point where the tolerance is satisfied, I won't get the error message? Hope anyone can help me on this. Thank you very much in advance. Best regards, Tony Example of the error message: Step 52, t= 2.5600000E+01, DT= 5.0000000E-02, C= 2.689 5.6747E+02 1.0237E+01 Solving for fluid T T T 52 PRES gmres: 29 9.7057E-07 1.0000E-06 1.5219E-03 1.0926E+00 2.1480E+00 F 52 Hmholtz VELX: 99 8.5954E-10 6.0898E-03 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELY 1.7088E-09 2.5049E-02 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4665E-09 2.5165E-02 1.0000E-09 L1/L2 DIV(V) : 2.7573E-18 9.1661E-04 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.7573E-18 9.1661E-04 52 2.5600E+01 1.0094E+01 Fluid done Step 53, t= 2.5650000E+01, DT= 5.0000000E-02, C= 2.697 5.7766E+02 1.0186E+01 Solving for fluid T T T 53 PRES gmres: 28 9.6076E-07 1.0000E-06 1.1322E-03 1.0558E+00 2.0542E+00 F 53 Hmholtz VELX: 98 9.4891E-10 5.8783E-03 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELY 1.6450E-09 2.4941E-02 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4160E-09 2.5049E-02 1.0000E-09 L1/L2 DIV(V) : 2.4695E-18 1.0845E-03 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.4695E-18 1.0845E-03 53 2.5650E+01 9.9794E+00 Fluid done -------------- 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 71, Issue 16 ********************************************* _______________________________________________ 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 Fri Jan 30 14:53:23 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Fri, 30 Jan 2015 20:53:23 +0000 Subject: [Nek5000-users] Error in Hmholtz solving In-Reply-To: References: Message-ID: Hi Paul, I've got it now. Thank you very much for the information. Best regards, Tony ________________________________________ From: nek5000-users-bounces at lists.mcs.anl.gov on behalf of nek5000-users-request at lists.mcs.anl.gov Sent: 30 January 2015 18:00 To: nek5000-users at lists.mcs.anl.gov Subject: Nek5000-users Digest, Vol 71, Issue 17 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: Error in Hmholtz solving (nek5000-users at lists.mcs.anl.gov) 2. Re: Error in Hmholtz solving (nek5000-users at lists.mcs.anl.gov) ---------------------------------------------------------------------- Message: 1 Date: Thu, 29 Jan 2015 19:44:28 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: Re: [Nek5000-users] Error in Hmholtz solving Message-ID: Content-Type: text/plain; charset="iso-8859-1" Hi Paul, Many thanks for your help. Yes, my simulation is scaled in convective time units. As you pointed out, the solutions are well converged. I just have a question about the projection and dealiasing. I'm still not sure when I need to use them in my simulations. Say I'm doing simulations without any models, do I need to turn on projection and dealiasing? I actually don't know how they will affect the solutions, would projection or dealiasing (off, old or new?) change the solutions? It would be great if you could briefly explain this bit to me. 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: 29 January 2015 18:00 To: nek5000-users at lists.mcs.anl.gov Subject: Nek5000-users Digest, Vol 71, 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. Error in Hmholtz solving (nek5000-users at lists.mcs.anl.gov) 2. Re: Error in Hmholtz solving (nek5000-users at lists.mcs.anl.gov) ---------------------------------------------------------------------- Message: 1 Date: Thu, 29 Jan 2015 15:50:43 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: [Nek5000-users] Error in Hmholtz solving Message-ID: Content-Type: text/plain; charset="iso-8859-1" Dear Neks, I'm getting the error message of '**ERROR**: Failed in HMHOLTZ: ' when running my simulation. I've checked the subroutine 'cggo' in hmholtz.f where the error message is printed. I understand that when the tolerance is not satisfied, the code would output this error message. But what I'm wondering is why the maximum iteration number is only 100. In the subroutine 'cggo', niter is equal to min(maxit,maxcg) where maxit=200 and maxcg=900. I can't see other places where maxit is restricted to only 100. So why the maximum iteration number is 100 when solving the Hmholtz equation? Also, is it right that if I increase the maximum number of iterations to certain point where the tolerance is satisfied, I won't get the error message? Hope anyone can help me on this. Thank you very much in advance. Best regards, Tony Example of the error message: Step 52, t= 2.5600000E+01, DT= 5.0000000E-02, C= 2.689 5.6747E+02 1.0237E+01 Solving for fluid T T T 52 PRES gmres: 29 9.7057E-07 1.0000E-06 1.5219E-03 1.0926E+00 2.1480E+00 F 52 Hmholtz VELX: 99 8.5954E-10 6.0898E-03 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELY 1.7088E-09 2.5049E-02 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4665E-09 2.5165E-02 1.0000E-09 L1/L2 DIV(V) : 2.7573E-18 9.1661E-04 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.7573E-18 9.1661E-04 52 2.5600E+01 1.0094E+01 Fluid done Step 53, t= 2.5650000E+01, DT= 5.0000000E-02, C= 2.697 5.7766E+02 1.0186E+01 Solving for fluid T T T 53 PRES gmres: 28 9.6076E-07 1.0000E-06 1.1322E-03 1.0558E+00 2.0542E+00 F 53 Hmholtz VELX: 98 9.4891E-10 5.8783E-03 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELY 1.6450E-09 2.4941E-02 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4160E-09 2.5049E-02 1.0000E-09 L1/L2 DIV(V) : 2.4695E-18 1.0845E-03 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.4695E-18 1.0845E-03 53 2.5650E+01 9.9794E+00 Fluid done -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Message: 2 Date: Thu, 29 Jan 2015 16:00:45 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: Re: [Nek5000-users] Error in Hmholtz solving Message-ID: Content-Type: text/plain; charset="iso-8859-1" Tony, It looks like your Helmholtz tolerance is too tight. If your problem is scaled in convective time units with characteristic length scale 1, timescale 1, and visc ~ 1/Re, then typical Helmholtz tolerances are around 1.e-7. You might also try turning on projection --- set p93=20 (and lmxprev=20 in SIZE), p94=5 and p95=5 in the rea file. No guarantees at this point because there seems to be something flaky in the projection routines these days under certain runtime combinations (e.g., I think splitting + heat transfer + velocity projection does not work)... It's on my todo list to clean these up. Note that your helmholtz solves are pretty well converged... residual is 1.7e-9, etc. Projection should easily put you over the top 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: Thursday, January 29, 2015 9:50 AM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Error in Hmholtz solving Dear Neks, I'm getting the error message of '**ERROR**: Failed in HMHOLTZ: ' when running my simulation. I've checked the subroutine 'cggo' in hmholtz.f where the error message is printed. I understand that when the tolerance is not satisfied, the code would output this error message. But what I'm wondering is why the maximum iteration number is only 100. In the subroutine 'cggo', niter is equal to min(maxit,maxcg) where maxit=200 and maxcg=900. I can't see other places where maxit is restricted to only 100. So why the maximum iteration number is 100 when solving the Hmholtz equation? Also, is it right that if I increase the maximum number of iterations to certain point where the tolerance is satisfied, I won't get the error message? Hope anyone can help me on this. Thank you very much in advance. Best regards, Tony Example of the error message: Step 52, t= 2.5600000E+01, DT= 5.0000000E-02, C= 2.689 5.6747E+02 1.0237E+01 Solving for fluid T T T 52 PRES gmres: 29 9.7057E-07 1.0000E-06 1.5219E-03 1.0926E+00 2.1480E+00 F 52 Hmholtz VELX: 99 8.5954E-10 6.0898E-03 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELY 1.7088E-09 2.5049E-02 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4665E-09 2.5165E-02 1.0000E-09 L1/L2 DIV(V) : 2.7573E-18 9.1661E-04 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.7573E-18 9.1661E-04 52 2.5600E+01 1.0094E+01 Fluid done Step 53, t= 2.5650000E+01, DT= 5.0000000E-02, C= 2.697 5.7766E+02 1.0186E+01 Solving for fluid T T T 53 PRES gmres: 28 9.6076E-07 1.0000E-06 1.1322E-03 1.0558E+00 2.0542E+00 F 53 Hmholtz VELX: 98 9.4891E-10 5.8783E-03 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELY 1.6450E-09 2.4941E-02 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4160E-09 2.5049E-02 1.0000E-09 L1/L2 DIV(V) : 2.4695E-18 1.0845E-03 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.4695E-18 1.0845E-03 53 2.5650E+01 9.9794E+00 Fluid done -------------- 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 71, Issue 16 ********************************************* ------------------------------ Message: 2 Date: Thu, 29 Jan 2015 20:05:04 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: Re: [Nek5000-users] Error in Hmholtz solving Message-ID: Content-Type: text/plain; charset="Windows-1252" Tony, Projection simply accelerates solution of the linear systems -- see: http://www.mcs.anl.gov/~fischer/pubhtml/project.pdf Dealiasing is invariably a good idea and not too expensive. It just ensure proper integration of the nonlinear terms and avoids some spurious eigenvalues. J. Malm, P. Schlatter, P. F. Fischer, and D. S. Henningson. Stabilization of the spectral element method in convection dominated flows by recovery of skew-symmetry. J. Sci. Comput., 57:254? 277, 2013. The real question is whether to use filtering or not... I generally do, and I do so by setting p103=0.01 and p101=0 in cases that are well-resolved. 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: Thursday, January 29, 2015 1:44 PM To: nek5000-users at lists.mcs.anl.gov Subject: Re: [Nek5000-users] Error in Hmholtz solving Hi Paul, Many thanks for your help. Yes, my simulation is scaled in convective time units. As you pointed out, the solutions are well converged. I just have a question about the projection and dealiasing. I'm still not sure when I need to use them in my simulations. Say I'm doing simulations without any models, do I need to turn on projection and dealiasing? I actually don't know how they will affect the solutions, would projection or dealiasing (off, old or new?) change the solutions? It would be great if you could briefly explain this bit to me. 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: 29 January 2015 18:00 To: nek5000-users at lists.mcs.anl.gov Subject: Nek5000-users Digest, Vol 71, 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. Error in Hmholtz solving (nek5000-users at lists.mcs.anl.gov) 2. Re: Error in Hmholtz solving (nek5000-users at lists.mcs.anl.gov) ---------------------------------------------------------------------- Message: 1 Date: Thu, 29 Jan 2015 15:50:43 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: [Nek5000-users] Error in Hmholtz solving Message-ID: Content-Type: text/plain; charset="iso-8859-1" Dear Neks, I'm getting the error message of '**ERROR**: Failed in HMHOLTZ: ' when running my simulation. I've checked the subroutine 'cggo' in hmholtz.f where the error message is printed. I understand that when the tolerance is not satisfied, the code would output this error message. But what I'm wondering is why the maximum iteration number is only 100. In the subroutine 'cggo', niter is equal to min(maxit,maxcg) where maxit=200 and maxcg=900. I can't see other places where maxit is restricted to only 100. So why the maximum iteration number is 100 when solving the Hmholtz equation? Also, is it right that if I increase the maximum number of iterations to certain point where the tolerance is satisfied, I won't get the error message? Hope anyone can help me on this. Thank you very much in advance. Best regards, Tony Example of the error message: Step 52, t= 2.5600000E+01, DT= 5.0000000E-02, C= 2.689 5.6747E+02 1.0237E+01 Solving for fluid T T T 52 PRES gmres: 29 9.7057E-07 1.0000E-06 1.5219E-03 1.0926E+00 2.1480E+00 F 52 Hmholtz VELX: 99 8.5954E-10 6.0898E-03 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELY 1.7088E-09 2.5049E-02 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4665E-09 2.5165E-02 1.0000E-09 L1/L2 DIV(V) : 2.7573E-18 9.1661E-04 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.7573E-18 9.1661E-04 52 2.5600E+01 1.0094E+01 Fluid done Step 53, t= 2.5650000E+01, DT= 5.0000000E-02, C= 2.697 5.7766E+02 1.0186E+01 Solving for fluid T T T 53 PRES gmres: 28 9.6076E-07 1.0000E-06 1.1322E-03 1.0558E+00 2.0542E+00 F 53 Hmholtz VELX: 98 9.4891E-10 5.8783E-03 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELY 1.6450E-09 2.4941E-02 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4160E-09 2.5049E-02 1.0000E-09 L1/L2 DIV(V) : 2.4695E-18 1.0845E-03 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.4695E-18 1.0845E-03 53 2.5650E+01 9.9794E+00 Fluid done -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Message: 2 Date: Thu, 29 Jan 2015 16:00:45 +0000 From: nek5000-users at lists.mcs.anl.gov To: "nek5000-users at lists.mcs.anl.gov" Subject: Re: [Nek5000-users] Error in Hmholtz solving Message-ID: Content-Type: text/plain; charset="iso-8859-1" Tony, It looks like your Helmholtz tolerance is too tight. If your problem is scaled in convective time units with characteristic length scale 1, timescale 1, and visc ~ 1/Re, then typical Helmholtz tolerances are around 1.e-7. You might also try turning on projection --- set p93=20 (and lmxprev=20 in SIZE), p94=5 and p95=5 in the rea file. No guarantees at this point because there seems to be something flaky in the projection routines these days under certain runtime combinations (e.g., I think splitting + heat transfer + velocity projection does not work)... It's on my todo list to clean these up. Note that your helmholtz solves are pretty well converged... residual is 1.7e-9, etc. Projection should easily put you over the top 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: Thursday, January 29, 2015 9:50 AM To: nek5000-users at lists.mcs.anl.gov Subject: [Nek5000-users] Error in Hmholtz solving Dear Neks, I'm getting the error message of '**ERROR**: Failed in HMHOLTZ: ' when running my simulation. I've checked the subroutine 'cggo' in hmholtz.f where the error message is printed. I understand that when the tolerance is not satisfied, the code would output this error message. But what I'm wondering is why the maximum iteration number is only 100. In the subroutine 'cggo', niter is equal to min(maxit,maxcg) where maxit=200 and maxcg=900. I can't see other places where maxit is restricted to only 100. So why the maximum iteration number is 100 when solving the Hmholtz equation? Also, is it right that if I increase the maximum number of iterations to certain point where the tolerance is satisfied, I won't get the error message? Hope anyone can help me on this. Thank you very much in advance. Best regards, Tony Example of the error message: Step 52, t= 2.5600000E+01, DT= 5.0000000E-02, C= 2.689 5.6747E+02 1.0237E+01 Solving for fluid T T T 52 PRES gmres: 29 9.7057E-07 1.0000E-06 1.5219E-03 1.0926E+00 2.1480E+00 F 52 Hmholtz VELX: 99 8.5954E-10 6.0898E-03 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELY 1.7088E-09 2.5049E-02 1.0000E-09 52 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4665E-09 2.5165E-02 1.0000E-09 L1/L2 DIV(V) : 2.7573E-18 9.1661E-04 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.7573E-18 9.1661E-04 52 2.5600E+01 1.0094E+01 Fluid done Step 53, t= 2.5650000E+01, DT= 5.0000000E-02, C= 2.697 5.7766E+02 1.0186E+01 Solving for fluid T T T 53 PRES gmres: 28 9.6076E-07 1.0000E-06 1.1322E-03 1.0558E+00 2.0542E+00 F 53 Hmholtz VELX: 98 9.4891E-10 5.8783E-03 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELY 1.6450E-09 2.4941E-02 1.0000E-09 53 100 **ERROR**: Failed in HMHOLTZ: VELZ 1.4160E-09 2.5049E-02 1.0000E-09 L1/L2 DIV(V) : 2.4695E-18 1.0845E-03 L1/L2 QTL : 0.0000E+00 0.0000E+00 L1/L2 DIV(V)-QTL: 2.4695E-18 1.0845E-03 53 2.5650E+01 9.9794E+00 Fluid done -------------- 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 71, Issue 16 ********************************************* _______________________________________________ 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 71, Issue 17 ********************************************* From nek5000-users at lists.mcs.anl.gov Sat Jan 31 23:05:07 2015 From: nek5000-users at lists.mcs.anl.gov (nek5000-users at lists.mcs.anl.gov) Date: Sun, 1 Feb 2015 05:05:07 +0000 Subject: [Nek5000-users] Changed mesh coordinates Message-ID: Dear Neks, I'm doing some data processing for my simulations in nek5000. I found for some elements, however, the coordinates are not exactly as specified. For example, I have an element whose vertices are as follows. ELEMENT 195 [ 1m] GROUP 0 0.000000E+00 -0.130526E+00 -0.129129E+00 0.000000E+00 0.100000E+01 0.991445E+00 0.979179E+00 0.987876E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00 -0.130526E+00 -0.129129E+00 0.000000E+00 0.100000E+01 0.991445E+00 0.979179E+00 0.987876E+00 0.263158E+00 0.263158E+00 0.263158E+00 0.263158E+00 When I used the following code to print the coordinates for this element, I got the values below. do e=1,nelv eg = lglel(e) if(eg .eq. 195) then write(*,*) xm1(1,1,1,e),ym1(1,1,1,e),zm1(1,1,1,e) write(*,*) cbc(1:6,e,1) end if end do 7.500381456730263E-017 1.00000000000000 0.000000000000000E+000 0.000000000000000E+000 W E E E P E Is there any reason why the x coordinate for the first element vertex is not exactly 0.0e0? Hope someone can help. Thank you very much. Best regards, Tony -------------- next part -------------- An HTML attachment was scrubbed... URL: