Users

users@lists.einsteintoolkit.org

3 participants
3984 discussions

Start a nNew thread

Einstein Toolkit Meeting
by Roland Haas 23 Sep '12

23 Sep '12

1 0

support_old_CalcTmunu_mechanism
by Frank Loeffler 23 Sep '12

23 Sep '12

Hi, Are there objections to change the default of TmunuBase::support_old_CalcTmunu_mechanism from "yes" to "no"? Frank

1 1

meeting minutes
by Roland Haas 22 Sep '12

22 Sep '12

Present were: Roland, Frank, Ian, Seth, Eloisa, Bruno, Erik, Peter, Tanja, Philipp ET goals/issues: * tickets/trac: - we seem to have a relatively large number of quite old tickets - seem to be doing ok in closing and fixing most current tickets - advertise existence of trac mailing list to mailing list. It is currently mentioned on http://einsteintoolkit.org/community/support/ but not in a very prominent place * there is a suggestion of moving away from using CCT accounts for trac/svn etc and instead generate independent ET accounts for this so as to not share a single password for shell accounts and less-critical accounts. It would also simplify creating trac-only accounts for users. Not conclusion was reached. * we want to keep the regular ET call at a once a week frequency, otherwise the danger is that participants might forget * using git repositories instead of svn - want to support mixed sets of repositories - general desire to support git through ET. No decision of whether we want to host these repositories ourself or use an external hoster. - most likely candidate to switch: GRHydro - if/when switching, provide tutorial or link to tutorial for new git users - expect no (or low) impact or read-only users that get a checkout via GetComponents - a desire to be able to use git-svn in GetComponents was voiced but not further discussed. Note that a ticket exists for this https://trac.einsteintoolkit.org/ticket/762 * we would like to expand the existing gallery of pictures https://secure.flickr.com/photos/48081610@N03 * movies would be very welcome * we would like to provide installation instructions for ET on "common" OSs: some variant of Linux (Ubuntu, Debian, RedHat?) and MacOSX detailing the packages to install and simfactory option files to use. Preferably starting from a "default" install of the OS. Some initial work is on https://docs.einsteintoolkit.org/et-docs/Simplified_Tutorial_for_New_Users * want to add gfortran to default simfactory options file (generic.cfg) * want Cactus build system to probe for gfortran since it is almost always present and needed for basic thorns, eg. Carpet. Cactus itself should not require gfortran though * Frank will not be able to host next weeks call, so Roland will do so Yours, Roland

1 2

EinsteinToolkit on the Janus supercomputer
by Bruno Giacomazzo 20 Sep '12

20 Sep '12

Hi, I have installed the EinsteinToolkit (Lovelace) on the Janus supercomputer here at the University of Colorado Boulder (https://www.rc.colorado.edu/resources/janus). Most of the testsuites pass (see also summary.log attached). The ones that give more problems are those from QuasiLocalMeasures which all fail because of segfault and I will try to understand why. I'd like to know if there is a "standard" benchmark parameter file that I can use to test the performance of GRHydro on Janus. Ideally it would be something simple (e.g., a TOV, so that I could use a very similar parfile to compare also GRHydro vs Whisky) and that you have used on other machines. I'm in particular interested in testing its weak scaling and possibly also its strong scaling. Thanks, Bruno Dr. Bruno Giacomazzo JILA - University of Colorado 440 UCB Boulder, CO 80309 USA Tel. : +1 303-492-5170 Fax : +1 303-492-5235 email : bruno.giacomazzo(a)jila.colorado.edu web: http://www.brunogiacomazzo.org ---------------------------------------------------------------------- There are only 10 types of people in the world: Those who understand binary, and those who don't ----------------------------------------------------------------------

2 3

Einstein Toolkit Meeting
by Frank Loeffler 17 Sep '12

17 Sep '12

Hi Please consider joining the weekly Einstein Toolkit phone call at 10 am US central time on Mondays. As usual, you can find instructions how to join on the following web site: http://einsteintoolkit.org/community/support/ I short: the number is (+1) 225-578-4942 or (+1) 866-573-0359 and the conference id is 118682#. Besides any development issues that might have come up last week I would like to brainstorm about the future direction of the toolkit. This is a good time to look back because officially the original grant will end soon. In no particular order and certainly not complete, think about topics of the following kind: - If you were the single person to change something within the ET, what would it be? - Funding for the next three years will be much less than what the project had so far, thus we need to concentrate on the most important things. What do you consider most important? - Are the weekly calls your tool of choice to coordinate development? - Are the services the toolkit provides sufficient? Should something in particular be changed / improved? - Should the toolkit have 'own' user profiles (not using the CCT logins)? - Should we investigate in non-svn repos at ET.org (specifically git)? - Should the web page stay plain html, editable by every developer? Some of these questions might be fast to answer, others maybe not and you might have some of your own. This is not meant to decide anything, but to get community input on what is needed/expected/wanted and we do this deliberately at the public call so that everyone can have a voice. Of course we would like to see emails as well in case you prefer that or you cannot join the call. Last week was busy: 16 Issues have been changed and are still open but 20 have been closed, e.g.: - The default relative tolerance of testsuites has been changed from 0 to 1.e-12 - The rsync option '-C' isn't used by simfactory anymore since it excluded copying some things users did in fact want to copy. - SummationByParts got improved OpenMP parallelization for Intel Compilers (because they still cannot handle workshares as they should). - LocalInterp got some OpenMP improvements. - Cactus now better enforces that arrangement and thorn names only contain characters from a whitelist. This is necessary because these names are used inside C code. - Carpet could in some cases interpolate data from buffer regions. This was now fixed. 9 Tickets are waiting for review at the moment [1]. Please see if you can help. Frank Loeffler [1] https://trac.einsteintoolkit.org/query?status=review&group=component&col=id… _______________________________________________ Users mailing list Users(a)einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users

1 0

unnecessary rebuilds
by Roland Haas 13 Sep '12

13 Sep '12

Hello all, I am still getting occasional unexplainable rebuilds of my Cactus configuration. Eg. just now I modified GRHydro/Con2Prim and the system recompiles McLachlan. Is anyone else experiencing these? And do you have an inkling as to what could cause them? Input would be welcome (and I will look into this but need some better starting point even if is just a list of situations where a rebuild was triggered). Yours, Roland -- My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from http://keys.gnupg.net.

2 4

visualizing field lines
by Roland Haas 13 Sep '12

13 Sep '12

Hello all, I tried to visualize the magnetic field lines for the rotor tests in VisIt. I can visualize it via a Bvec vector plot but when trying the field lines plot VisIt finds and error that no plot was produced. Did anyone ever manage to produce a field lines plot? You an actually find the data in the svn repository for the paper: https://svn.einsteintoolkit.org/documents/Paper_EinsteinToolkit_GRMHD/rotor… Yours, Roland -- My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from http://keys.gnupg.net.

3 4

Re: [Users] [Commits] [svn:einsteintoolkit] GRHydro/trunk/ (Rev. 421)
by Erik Schnetter 12 Sep '12

12 Sep '12

I accidentally committed more changes than intended. The only other relevant change is to make Y_e_con_rhs public instead of private. This is in line with the other rhs variables, and is needed for refluxing. -erik On Wed, Sep 12, 2012 at 2:00 PM, <schnetter(a)cct.lsu.edu> wrote: > User: eschnett > Date: 2012/09/12 01:00 PM > > Modified: > /trunk/ > interface.ccl > /trunk/src/ > GRHydro_Con2Prim.F90, GRHydro_TVDReconstruct.F90, GRHydro_Tmunu.F90, > GRHydro_TmunuM.F90 > > Log: > Correct format specifier > > File Changes: > > Directory: /trunk/src/ > ====================== > > File [modified]: GRHydro_Con2Prim.F90 > Delta lines: +1 -1 > =================================================================== > --- trunk/src/GRHydro_Con2Prim.F90 2012-09-04 01:20:40 UTC (rev 420) > +++ trunk/src/GRHydro_Con2Prim.F90 2012-09-12 18:00:56 UTC (rev 421) > @@ -859,7 +859,7 @@ > if(nf.gt.nfudgemax) then > if(GRHydro_c2p_reset_eps_tau_hot_eos.ne.1) then > call CCTK_WARN(failinfomode,"EOS error in c2p 0: > injected heat too many times") > - write(warnline,"(i8.4i5,1P10E15.6)") > cctk_iteration,GRHydro_reflevel,ii,jj,kk,x,y,z > + write(warnline,"(i8,4i5,1P10E15.6)") > cctk_iteration,GRHydro_reflevel,ii,jj,kk,x,y,z > call CCTK_WARN(failinfomode,warnline) > write(warnline,"(1P10E15.6)") rho,epsilon,temp,ye > call CCTK_WARN(failinfomode,warnline) > > File [modified]: GRHydro_TVDReconstruct.F90 > Delta lines: +1 -1 > =================================================================== > --- trunk/src/GRHydro_TVDReconstruct.F90 2012-09-04 01:20:40 UTC > (rev 420) > +++ trunk/src/GRHydro_TVDReconstruct.F90 2012-09-12 18:00:56 UTC > (rev 421) > @@ -115,7 +115,7 @@ > CCTK_REAL,intent(IN)::a_in,b_in > CCTK_REAL::minmod_result > > - > minmod_result=0.5D0*(sign(1.0D0,a_in)+sign(1.0D0,b_in))*min(abs(a_in),abs(b_in)) > + > minmod_result=0.5D0*(sign(1.0d0,a_in)+sign(1.0d0,b_in))*min(abs(a_in),abs(b_in)) > end function minmod_func > > end subroutine tvdreconstruct > > File [modified]: GRHydro_Tmunu.F90 > Delta lines: +1 -1 > =================================================================== > --- trunk/src/GRHydro_Tmunu.F90 2012-09-04 01:20:40 UTC (rev 420) > +++ trunk/src/GRHydro_Tmunu.F90 2012-09-12 18:00:56 UTC (rev 421) > @@ -105,7 +105,7 @@ > !!$ Calculate the specific relativistic enthalpy times rho times the > !!$ square of the lorentz factor. > > - rhoenthalpy = w_lorentz(i,j,k)**2*(rho(i,j,k)*(1.0D0 + > eps(i,j,k)) + press(i,j,k)) > + rhoenthalpy = w_lorentz(i,j,k)**2*(rho(i,j,k)*(1.0d0 + > eps(i,j,k)) + press(i,j,k)) > > !!$ Calculate lower components of 4-velocity (without the Lorent > factor). > > > File [modified]: GRHydro_TmunuM.F90 > Delta lines: +1 -1 > =================================================================== > --- trunk/src/GRHydro_TmunuM.F90 2012-09-04 01:20:40 UTC (rev 420) > +++ trunk/src/GRHydro_TmunuM.F90 2012-09-12 18:00:56 UTC (rev 421) > @@ -122,7 +122,7 @@ > !!$ Calculate the specific relativistic enthalpy times rho + the > mag. field contribution times the > !!$ square of the lorentz factor. > > - rhohstarw2 = w_lorentz(i,j,k)**2*(rho(i,j,k)*(1.0D0 + > eps(i,j,k)) + press(i,j,k) + b2) > + rhohstarw2 = w_lorentz(i,j,k)**2*(rho(i,j,k)*(1.0d0 + > eps(i,j,k)) + press(i,j,k) + b2) > pstar = press(i,j,k)+0.5d0*b2 > > !!$ Calculate lower components of 4-velocity (without the Lorent > factor). > > Directory: /trunk/ > ================== > > File [modified]: interface.ccl > Delta lines: +1 -1 > =================================================================== > --- trunk/interface.ccl 2012-09-04 01:20:40 UTC (rev 420) > +++ trunk/interface.ccl 2012-09-12 18:00:56 UTC (rev 421) > @@ -611,8 +611,8 @@ > fs_alpha5 > } "Maximum characteristic speeds for the flux splitting" > > +public: > real Y_e_con_rhs TYPE=GF tags='Prolongation="None" checkpoint="no"' "RHS > for the electron fraction" > -public: > real Y_e_con_flux TYPE=GF tags='Prolongation="None" checkpoint="no"' > "Flux for the electron fraction" > private: > real Y_e_plus TYPE=GF tags='Prolongation="None" checkpoint="no"' "Plus > state for the electron fraction" > > _______________________________________________ > Commits mailing list > Commits(a)einsteintoolkit.org > http://lists.einsteintoolkit.org/mailman/listinfo/commits > -- Erik Schnetter <schnetter(a)cct.lsu.edu> http://www.perimeterinstitute.ca/personal/eschnetter/

1 0

poly_gamma_ini in EOS_Omni
by Frank Loeffler 11 Sep '12

11 Sep '12

Hi, It has been proposed [1] to remove the EOS_Omni parameter poly_gamma_ini. Please comment if you use it, otherwise it will be removed. Frank Löffler [1] https://trac.einsteintoolkit.org/ticket/1070

1 0

meeting minutes
by Roland Haas 10 Sep '12

10 Sep '12

Present were: Ian, Eloisa, Seth, Frank, Roland, Erik, Steve, Peter, Tanja, Yosef, Philipp thorns in incoming: * SphericalSlice - no progress yet - used by worltube for CCE ID extraction - needs documentation, tests - talk to author (Christian Reisswig) * SphericalHarmonicReconGen - lacking test case, docs - Roland will try to get test data for data formats * Outflow: - please register interest ET grant work process: Please think about and send out opinions about: * what do you think of the way that the CIGR project evolved during the lifetime of the grant, in particular - collaboration among member institutions - code management: -- fully open vs. development in privacy -- (development) branches? -- private clones (reasons for creating them)? -- proliferation of branches for individual projects -- make more of the responsibility allocation/funding public (this concerns the PIs)? Please make an effort to form up a plan about these: * what do you consider important goals for the ET in the next few years? * who is working on these projects? Yours, Roland

2 2

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

Users