Users

users@lists.einsteintoolkit.org

3 participants
3984 discussions

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

Einstein Toolkit Meeting Reminder
by Frank Loeffler 15 Nov '16

15 Nov '16

1 2

Calling BLAS from Fortran
by Ian Hinder 14 Nov '16

14 Nov '16

Hi, What is the correct way to call a BLAS routine from a Fortran Cactus thorn? I am trying to compile the LlamaWaveToy, and it does this: subroutine LWT_min_spacing (CCTK_ARGUMENTS) use lapack ... call syev ('N', 'U', 3, gg, 3, ev, work, lwork, info) and at link time, I get /home/ianhin/Cactus/EinsteinToolkitGit/configs/etllama/lib/libthorn_LlamaWaveToy.a(min_spacing.F90.o): In function `lwt_min_spacing': /home/ianhin/Cactus/EinsteinToolkitGit/arrangements/Llama/LlamaWaveToy/src/min_spacing.F90:38: undefined reference to `syev_' Do I need to do something special to make the symbol found, or to convert Fortran to C names, etc? -- Ian Hinder http://members.aei.mpg.de/ianhin

3 2

Information about grid decomposition
by breu＠th.physik.uni-frankfurt.de 10 Nov '16

10 Nov '16

Dear users of the Einstein Toolkit, I have a question: I would like to integrate the geodesic equations for a large quantity of particles in parallel (e.g. one million). To save computational time, I want to make sure that the arrays containing the particle data are handled by the same process that takes care of the corresponding grid patch on which the particles are moving and only communicate particle data if a particle moves to another grid patch. Since I only want a process to communicate with the processes that contain the adjacent grid patches, can I access information on which process owns which grid patch and if yes, how? So far I have only found functions that return the upper and lower bounds for the local process, but each process would need to know where all the rest of the grid is. Kind regards, Cosima Breu

3 3

ET release fails 4 tests
by Ian Hinder 09 Nov '16

09 Nov '16

Hi, I reran the "EinsteinToolkitReleased" Jenkins job today, and it has 4 failing tests. The code has not changed since the previous run on 7th October (when all tests passed), but the build and test node is using a different OS and CPU. The change in CPU was unavoidable as the old node is no longer available. The change in OS was somewhat unintentional, but I suspect the problem comes from the CPU. I believe that this was also observed for trunk; is that correct? 3 out of 4 failures are related to NaNs. The 4th is not. https://build-test.barrywardell.net/job/EinsteinToolkitReleased/107/testRep… -- Ian Hinder http://members.aei.mpg.de/ianhin

1 0

meeting minutes for 2016-11-07
by Roland Haas 07 Nov '16

07 Nov '16

Present: Eloisa, Frank, Haris, Vassili, Ian, Josh, Steve, Roland, Erik ET release: * planned for end of November * run test suite on your local workstations to make sure we cover all flavors of Linux, OSX etc. * Roland to run on BW, Frank to create a list of desired machines and contact possible runners * Roland to add the hard-link fixes back to formaline to improve build speed Fortran code in GRHydro: * announce in release that Fortran code will go away in the next release. Point out the CXX code has been the default for at least one release. Give users opportunity to speak up if it should be retained. Llama into the ET: * needs push: Ian, Peter volunteers and will recruit people * needs check on documentation, needs examples and test cases 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 07 Nov '16

07 Nov '16

1 1

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

Users