Users November 2016

users@lists.einsteintoolkit.org

10 participants
17 discussions

Re: [Users] [ET Trac] [Einstein Toolkit] #1326: running loopcontrol on strange number of threads fails
by Erik Schnetter 11 Apr '19

11 Apr '19

Do you have a backtrace? -erik On Sunday, March 22, 2015, Einstein Toolkit < trac-noreply(a)einsteintoolkit.org> wrote: > #1326: running loopcontrol on strange number of threads fails > ------------------------------------+--------------------------------------- > Reporter: rhaas | Owner: > Type: defect | Status: new > Priority: minor | Milestone: > Component: EinsteinToolkit thorn | Version: > Resolution: | Keywords: LoopControl > ------------------------------------+--------------------------------------- > > Comment (by rhaas): > > This still happens even with current (Sun Mar 22 18:46:21 CET 2015) trunk, > though failure looks a bit different now: > {{{ > actus_sim: > /data/rhaas/postdoc/gr/ET_trunk/configs/sim/build/LoopControl/loopcontrol.cc:312: > T {anonymous}::divexact(T, T) [with T = int]: Assertion `i % j == 0' > failed > }}} > > -- > Ticket URL: <https://trac.einsteintoolkit.org/ticket/1326#comment:1> > Einstein Toolkit <http://einsteintoolkit.org> > The Einstein Toolkit > _______________________________________________ > Trac mailing list > Trac(a)einsteintoolkit.org > http://lists.einsteintoolkit.org/mailman/listinfo/trac > -- Erik Schnetter <schnetter(a)cct.lsu.edu> http://www.perimeterinstitute.ca/personal/eschnetter/

3 2

release progress meeting
by Frank Loeffler 30 Nov '16

30 Nov '16

Hi, A few of us agreed to have a short meeting about the release progress this (US) morning at 8:30am central time. I'd like for us to test a mumble sever at CCT for this. Don't expect a finished setup though - this is just a first try if connections and clients work. In case this doesn't work, I have opened a hangout as well (link at the end). mumble is available for virtually every platform; you most likely find it in your favorite distribution or app store. Just in case, here is it's home page: https://wiki.mumble.info/wiki/Main_Page The connection info you'll need is: mumble.cct.lsu.edu port 64738 (the default) (use your mumble client, not your web browser) You don't need a password to connect, but mumble might ask you to create a user certificate nevertheless when you start it the first time. Frank hangout link, in case there are problems with mumble: https://hangouts.google.com/call/yqgdbg336bbrnjcedebet3uwrie

1 0

Fwd: EinsteinToolkit - Build #884 - Still Unstable - 362/419 tests passed
by Ian Hinder 29 Nov '16

29 Nov '16

Hi, A recent commit caused a huge number of test failures. I looked at one, and it seemed to be caused by a floating point exception. We shouldn't get these, since we are supposed to disable them in favour of producing NaNs. The only change that looks like it might have caused this is to the NaNChecker. Roland, could your changes have had some unintended consequence? PS: don't update your checkout from trunk at the moment if you want a working toolkit :) Begin forwarded message: > From: Einstein Toolkit Jenkins <jenkins(a)build-test.barrywardell.net> > Subject: EinsteinToolkit - Build #884 - Still Unstable - 362/419 tests passed > Date: 28 November 2016 23:11:28 CET > To: ian.hinder(a)aei.mpg.de > > BUILD UNSTABLE > > Build URL: https://build-test.barrywardell.net/job/EinsteinToolkit/884/ > Project: EinsteinToolkit > Date of build: Mon, 28 Nov 2016 21:24:36 +0000 > Build duration: 46 min > > Changes > > Revision: 7e2d8addfd85e59a2ec06fffe31adc7c6e7f2f9e > Author: Einstein Toolkit Git Server > Log: > Updated submodules: > > * simfactory 2392b21...181074d (25): > > tezpur: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > sunnyvale: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > saw: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > s-kraken: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > requin: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > philip: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > pandora: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > osx-macports: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > mp2: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > intrepid: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > guillimin: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > gpc: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > gordon: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > generic: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > generic-mpi: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > gelileo: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > fermi: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > debian: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > datura: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > datura: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > datura: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > compute: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > carver: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > angel: add CACTUS_NUM_PROCS and CACTUS_NUM_THREADS to runscript > > osx-homebrew.cfg: add subversion to list of packages to install > Revision: f10df8c6c59d8aa146edef695870a591aad8fe45 > Author: Einstein Toolkit Git Server > Log: > Updated submodules: > > * arrangements/CactusUtils a810148...9583b71 (3): > > Formaline: comment on why tarballs are not included in git > > Formaline: only update files in thorns that changed > > Formaline: attempt hardlink of target files before resorting to update-index > > * simfactory 181074d...191e5a6 (1): > > changes to smic > Revision: 3095e456ab7d0ec55edf462e1b17fbfd6a8fe4f0 > Author: Einstein Toolkit Git Server > Log: > Updated submodules: > > * arrangements/CactusUtils 9583b71...05b9f1b (3): > > NaNChecker: always schedule routines > > NaNChecker: use internally generated NaNs in test case > > NaNChecker: add routine to set up test case > > Tests > > CarpetEvolutionMask.CarpetEvolutionMask_test/1procs New failure > CarpetEvolutionMask.CarpetEvolutionMask_test_off/1procs New failure > CarpetInterp.waveinterp-1p/1procs New failure > CarpetInterp.waveinterp-2p/2procs New failure > Coordinates.test7patch/2procs New failure > GRHydro.GRHydro_test_shock/1procs New failure > GRHydro.GRHydro_test_shock/2procs New failure > GRHydro.GRHydro_test_shock_hllc/1procs New failure > GRHydro.GRHydro_test_shock_hllc/2procs New failure > GRHydro.GRHydro_test_shock_mp5/1procs New failure > GRHydro.GRHydro_test_shock_mp5/2procs New failure > GRHydro.GRHydro_test_shock_ppm/1procs New failure > GRHydro.GRHydro_test_shock_ppm/2procs New failure > GRHydro.GRHydro_test_shock_weno/1procs Failed for 21 builds since build #864 on Fri Nov 04 16:06:26 UTC 2016 > GRHydro.GRHydro_test_shock_weno/2procs Failed for 21 builds since build #864 on Fri Nov 04 16:06:26 UTC 2016 > Hydro_InitExcision.diag_flip_pugh_eno/1procs New failure > Hydro_InitExcision.diag_flip_pugh_eno/2procs New failure > Hydro_InitExcision.diag_pugh_eno/1procs New failure > Hydro_InitExcision.diag_pugh_eno/2procs New failure > Hydro_InitExcision.sphere_pugh_ppm/1procs New failure > Hydro_InitExcision.sphere_pugh_ppm/2procs New failure > Hydro_InitExcision.x_flip_pugh_eno/1procs New failure > Hydro_InitExcision.x_flip_pugh_eno/2procs New failure > Hydro_InitExcision.x_pugh_eno/1procs New failure > Hydro_InitExcision.x_pugh_eno/2procs New failure > LoopControl.test-minkowski-carpet-1lev/2procs New failure > ML_BSSN_Test.ML_BSSN_EE_sgw3d/1procs New failure > ML_BSSN_Test.ML_BSSN_EE_sgw3d/2procs New failure > ML_BSSN_Test.ML_BSSN_EE_sgw3d_rhs/1procs New failure > ML_BSSN_Test.ML_BSSN_EE_sgw3d_rhs/2procs New failure > ML_BSSN_Test.ML_BSSN_NewRad/1procs New failure > ML_BSSN_Test.ML_BSSN_NewRad/2procs New failure > ML_BSSN_Test.ML_BSSN_O8_sgw3d/1procs New failure > ML_BSSN_Test.ML_BSSN_O8_sgw3d/2procs New failure > ML_BSSN_Test.ML_BSSN_sgw3d/1procs New failure > ML_BSSN_Test.ML_BSSN_sgw3d/2procs New failure > ML_BSSN_Test.ML_BSSN_sgw3d_harmonic/1procs New failure > ML_BSSN_Test.ML_BSSN_sgw3d_harmonic/2procs New failure > ML_BSSN_Test.ML_BSSN_sgw3d_harmonic_phi/1procs New failure > ML_BSSN_Test.ML_BSSN_sgw3d_harmonic_phi/2procs New failure > ML_BSSN_Test.ML_BSSN_sgw3d_rhs/1procs New failure > ML_BSSN_Test.ML_BSSN_sgw3d_rhs/2procs New failure > ML_CCZ4_Test.ML_CCZ4_sgw3d/1procs New failure > ML_CCZ4_Test.ML_CCZ4_sgw3d/2procs New failure > ML_CCZ4_Test.ML_CCZ4_sgw3d_rhs/1procs New failure > ML_CCZ4_Test.ML_CCZ4_sgw3d_rhs/2procs New failure > NaNChecker.nancount/1procs Failed for 21 builds since build #864 on Fri Nov 04 16:06:26 UTC 2016 > NaNChecker.nancount/2procs Failed for 21 builds since build #864 on Fri Nov 04 16:06:26 UTC 2016 > PeriodicCarpet.ml-gw1d-small-amr/2procs New failure > PeriodicCarpet.ml-gw1d-small/2procs New failure > RotatingSymmetry180.KerrSchild-rotating-180-EE/2procs New failure > RotatingSymmetry90.KerrSchild-rotating-90-EE/2procs New failure > Trigger.trigger/1procs New failure > Trigger.trigger/2procs New failure > TwoPunctures.twopunctures_carpet/2procs New failure > WaveMoL.gaussian/1procs New failure > WaveMoL.gaussian/2procs New failure > 362 tests passed > > Artifacts > > build__1_4.log > build__2_2.log > > -- Ian Hinder http://members.aei.mpg.de/ianhin

3 4

Machines used with the Einstein Toolkit
by Ian Hinder 29 Nov '16

29 Nov '16

Hi, For the upcoming release, we have only a small number of machines on which the tests have been run (http://einsteintoolkit.org/release-info/parse_testsuite_results.php). Usually we have many more. It would be very helpful to find out what machines people are actually wanting to use with the toolkit, so that we can focus testing on machines that people actually care about. Please could you let us know which machines are important for you? -- Ian Hinder http://members.aei.mpg.de/ianhin

6 10

Remote Mini-Workshop: FunHPC and friends
by Erik Schnetter 29 Nov '16

29 Nov '16

A few weeks ago we tried to schedule a remote mini-workshop on FunHPC. This failed for several reasons (I caught a cold, was travelling, etc.), but now it's time to revive this. I've created a wiki page at < https://docs.einsteintoolkit.org/et-docs/Remote_Mini-Workshop_Series>. I've suggested a few more topics, not just FunHPC. If you're interested, please go there and vote. I suggest Wed, Dec. 7 (9:00 EST) as meeting time, as per continuation of our earlier Doodle poll. -erik -- Erik Schnetter <schnetter(a)gmail.com> http://www.perimeterinstitute.ca/personal/eschnetter/

1 0

Re: [Users] [Einstein Toolkit] #1884: provide support for multi-model runs in thorn MPI
by Roland Haas 28 Nov '16

28 Nov '16

Hello all, do you want all of these applied before the release or only some (not sure which ones had the release milestone) or all but the one that explicitly says "after the release"? Yours, Roland > #1884: provide support for multi-model runs in thorn MPI > ------------------------------------+--------------------------------------- > Reporter: rhaas | Owner: > Type: enhancement | Status: review > Priority: minor | Milestone: > Component: EinsteinToolkit thorn | Version: development version > Resolution: | Keywords: MPI > ------------------------------------+--------------------------------------- > > Comment (by rhaas): > > I am no longer sure. > > For C, the functionality is the identical between this and Carpet I think. > The code would only collect the functionality on one spot rather than in > each driver. > > I think for Fortran, however, Carpet's aliased functions won't work since > there is no way to dereference the returned CCTK_POINTER_TO_CONST. The > returned pointer is a pointer to the C MPI communicator struct, which (as > far as I know) does not have to be bit-identical to Fortran's > MPI_Communicator type (which is INT). > > Having said all of this, the original usage case I had for this no longer > exists (I think it had to do with running PUGH under LoadLeveler or > something similar, I have PUGH/Carpet patches that would make user of this > code to provide the aliased functions). The implementation of the Reset > functions is also somewhat ugly. > -- 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.

1 0

meeting minutes for 2016-11-28
by Roland Haas 28 Nov '16

28 Nov '16

Present: Erik, Frank, Harry, Ian, Josh, Steve, Roland, Zach Failing test machines: * Ubuntu-lts: Roland to check if MPI processors are incorrect, set CACTUS_NUM_PROCS in generic.run * OSX Sierra: Roland to check if MPI processors are incorrect, set CACTUS_NUM_PROCS in generic.run * Comet & Gordon: Roland to check for common failures, and possibly hand to Ian or Frank if too many fail Failing test suites: * Dissipation: Peter will check * RotatingSymmetry90 + 180: Frank and Peter will check * WaveToyOpenCL fails on the two machine where it fails, Steve will try on Spine to see if CPU or GPU implementation failed Updated ET page: * Ian suggests to reduce content on first page * Roland would like to have material that is less used to be on separate pages, would like to have a "Download" link very visible and standing out on the front page * Roland would suggest something eye catching to capture interest * Roland would like to have most referred to items (Download, Help) accessible from the first page * would like to make sure that our website agrees with current best practises * sample is here http://einsteintoolkit.org/new/index.html * local preview is possible, content is plain html with some javascript in a mostly centralized location Formaline slowness: * Erik will review the patches in # 1982 Llama: * https://docs.einsteintoolkit.org/et-docs/Adding_Llama_to_the_ET * Barry worked through documentation * some bug left in Llama wavetoy EinsteinExamples: * will move gallery parfiles to EinsteinExamples and link to the EinsteinExamples parfiles on github * ComputeCanada is looking for cool astro data, which matches with what Eloisa, Ian, Barry wanted to do anyway for the GW150914 simulation. Data size would be ~TB. Name: * Cecilia Payne-Gaposchkin https://en.wikipedia.org/wiki/Cecilia_Payne-Gaposchkin 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.

1 0

Einstein Toolkit Meeting Reminder
by Frank Loeffler 28 Nov '16

28 Nov '16

Hi Please consider joining the weekly Einstein Toolkit phone call at 9:30 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/ We use a Google hangout, with an URL announced on this list before the meeting. Frank Loeffler

1 1

Einstein Toolkit and geodesic flow
by Charalampos Markakis 22 Nov '16

22 Nov '16

Dear Cosima, Regarding solving for thousands of geodesics, a faster alternative would be to solve them as a PDE, i.e. as a dust fluid. This is analogous to solving the Burgers equation (the pressureless Euler equation) for nonrelativistic flows. The Einstein Toolkit is naturally designed to handle such scenarios. But the details depend on what you are trying to simulate. For example, for timelike geodesics, this could be done by evolving eq. (10) in https://arxiv.org/abs/1410.7777 with p_a given by eq. (8) and H given by eq. (16), setting the specific enthalpy to h=1. Some caveats: 1) For null geodesics, you can do something similar, but need a different foliation. 2) If you try to use the Valencia formulation to evolve dust, you must evolve the continuity equation for the density \rho, along with the Euler equation for the velocity. This system is ill-posed and unstable for dust (pressure = 0), because you are evolving an equation that is no-longer needed: One needs to drop the continuity equation, eliminate \rho completely from the equations, and evolve only 3 equations for the spatial components of the 4-velocity. Then the system is well-posed (strongly hyperbolic), just like the Burgers equation (whose relativistic generalization you are solving). Eq. (10) in the above paper does this. 3) As Roland Haas pointed out, because you are evolving dust as a pressureless fluid, even though the flow is geodesic, caustics or shocks may form. If that happens, the way to handle it depends on what you are trying to simulate with geodesics. If you need to simulate the actual geodesics, and not dust, you can use the method of characteristics and disregard shock solutions. (For null geodesics you won't have this problem.) 4) If you need to know where each geodesic particle is at a given time, along with eq. (10) above, you can solve the other Hamilton equation for the position x^a conjugate to p_a, $ dx^a/dt=\partial H / \partial p_a $, written in Eulerian PDE form. This evolves a 'label' of the initial position of each particle. For any clarification, please feel free to contact me or Roland Haas. Best regards, Haris Markakis ________________________________________________ Charalampos Markakis Postdoctoral Research Associate National Center for Supercomputing Applications University of Illinois at Urbana-Champaign 1205 West Clark St, Room 4022, MC-257, Urbana, IL 61801 markakis(a)illinois.edu ________________________________________________ On 21 November 2016 at 10:56, Frank Loeffler <knarf(a)cct.lsu.edu> wrote: > On Mon, Nov 21, 2016 at 12:26:59AM -0600, Frank Loeffler wrote: > >> Please consider joining the weekly Einstein Toolkit phone call at >> > > The following group called in: > > Cosima (and a few others from the Frankfurt group; > I don't remember your names, sorry) > Erik > Frank > Harry > Ian > Peter > Roberto > Vassili > Yosef > Zach > > One topic was how to optimized a geodesics iteration thorn written by Ian > Hinder et.al., and now used by the Frankfurt group. The main problem is > that it was written with a few geodesics in mind, but should now be used > for millions. Erik strongly suggests to profile what exactly is slow before > starting to heavily optimized what you might think is slow. Assuming it is > communication, he gave a few hints which functions might help > (locatePosition?). > > Connected to this it was mentioned that integrating GAs is currently > broken if you have more than one component per process, as they are > currently integrated (by MoL) fore each component. Current workaround: use > more processes (to get 1 component per process). > > Release related issues: > > - Llama was added to the thornlist, including testsuites > - initial test results look quite good > (a few failures from usual suspects) > - Intel 17 compiler crashes on bbox.cc in Carpet. > - current workaround: lower optimization of that file to -O1 > - maybe better: use bbox1-class instead of newer bbox2 class. > however: you will have worse scaling for large runs > - current machines showing this problem: marconi, stampede-knl > - Only a few machines saw testing so far. We need more! > > A first shot at a new design of the ET pages is online: > > http://einsteintoolkit.org/new/ > > These aren't polished yet, but should give a first impression. They are > responsive, i.e., they should look presentable also on small screens > (phones). When you test them, please use different browsers, different font > sizes (zoom levels), different window sized, and different devices. > > Please send feedback either to the list, or directly to Frank < > knarf(a)cct.lsu.edu>. > > Frank > > > _______________________________________________ > Users mailing list > Users(a)einsteintoolkit.org > http://lists.einsteintoolkit.org/mailman/listinfo/users > >

1 0

Einstein Toolkit Meeting Reminder
by Frank Loeffler 21 Nov '16

21 Nov '16

3 5

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

Users November 2016