Users

users@lists.einsteintoolkit.org

3 participants
3984 discussions

meeting minutes for 2017-05-08
by Roland Haas 08 May '17

08 May '17

Present: Bhavesh, Eloisa, Frank. Gabrielle, Peter, Steve, Gela Hammerling, Ian, Roberto, Vassili, Zach, Roland, Qian Report on ET workshop planning: * planned for week starting July 31st, 3 days of school, 2 days of workshop * themes not fully decided yet, will sent out poll soon as possible (today) to query prior knowledge, which will let us asses where to pick people up. Topics will be a mix of lectures on NR and numerical methods plus hands on session incl. visualization, ET+LIGO * should be able to broadcast and record the lectures since NCSA has rooms equipped with cameras and trained staff to handle * workshop will likely dedicate one day to actual work and collaboration, will likely ask people to announce what they would like to work on at the time or registration * suggestion to ask for example Bill Gropp on a keynote like talk for the workshop ET release: * postponed a bit this year * request for review of new LORENE thorn * Piraha parser needs to have the tickets in review applied, debug output removed * failing test suites must be fixed, we will not release with failing tests: https://build-test.barrywardell.net/job/EinsteinToolkit/lastCompletedBuild/… * Frank will try and look at the GRHydro test * Roland will look at SphericalHarmonicReconGen GiRaFFE: * Zach gave a quick overview about it, https://arxiv.org/pdf/1704.00599.pdf * not for inclusion in this release yet Survey on ET data vault: https://docs.google.com/forms/d/e/1FAIpQLSdLrcXVJXKZBv3PAIaXtgj6YMylPVxwx-c… * there was discussion on the proper licenses to offer and how indexing by search engines would be implemented 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://pgp.mit.edu .

3 2

Einstein Toolkit Meeting Reminder
by Frank Loeffler 08 May '17

08 May '17

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 about 15 minutes before the meeting, and will also posted on https://docs.einsteintoolkit.org/et-docs/Main_Page Tomorrow, representatives from the NSF grant sites will give a short summary (not more than 5 minutes each) of the current states of their sub-projects. We will continue to do this roughly monthly from now on. Other projects involving the Einstein Toolkit are encouraged to chip in and report about their progress as well. Another topic of discussion will be the upcoming Einstein Toolkit workshop, likely to happen end of July at NCSA. This is only in the early stages of planning, so input is now especially valuable. Frank Löffler

4 6

Benchmarking
by Khamesra, Bhavesh 08 May '17

08 May '17

Hi, I have sent the pull request with the optionlist for Stampede - KNL on Bitbucket simfactory repo. I have tested this with a couple of thornlists including the einsteintoolkit.th and GW150914.th. This is still in experimental stage and so would be great if someone could also test it. Working on benchmarking the performance on Stampede KNL, I was able to do some test runs using the GW150914 simulation. However, I have been running into some issues with it. 1. I tried running QC0 simulation on both Stampede SandyBridge and KNL. While it runs fine on Stampede but it crashes on KNL with this error - while executing schedule bin BoundaryConditions, routine RotatingSymmetry180::Rot180_ApplyBC in thorn RotatingSymmetry180, file /work/04082/tg833814/Cactus_ETK_dev/arrangements/CactusNumerical/RotatingSymmetry180/src/rotatingsymmetry180.c:460: -> TAT/Slab can only be used if there is a single local component per MPI process TACC: MPI job exited with code: 134 I looked up at previous tickets and found the solution to increase the number of cores. But if the same simulation can be run on stampede on 64 cores, why does it require higher number of cores on KNL? Or is it some other issue? 2. I was able to run GW150914 on development queue (68 cores) and the speeds on Stampede were around 12.9M while that on KNL goes around 2.4M. To understand the reason for such small speeds, I tried running this on higher number of cores on Stampede (128) and it runs at speed of around 20.9M (tested the run for 12 hours). However, on doing the same in normal queue in KNL, the simulation crashes after a couple of iterations on KNL with some segmentation fault error. Also, before crashing, the speed on KNL is around 4.2M. I have attached the error file of the simulation. Could someone please look at this? Let me know if you need any other information. Thanks ............................. Bhavesh Khamesra Graduate Student Centre of Relativistic Astrophysics Georgia Institute of Technology

4 13

SimulationTools Frequency
by Gwyneth Allwright 06 May '17

06 May '17

Hi Ian, I've been finding SimulationTools very useful for looking at the results of my Cactus simulations. I just have a question about the function called "Frequency", which is used in the context of gravitational waveforms. How is this frequency defined? Is it an angular frequency? I don't know how to interpret the negative values I'm getting. Best, Gwyneth

2 2

Error about a shared object file
by Hiro.Kig 03 May '17

03 May '17

Dear all I am Hiromitsu Kigure. This is the first time I send an e-mail to this mailing list. I am sending you this email to seek your advice. I use Einstein Toolkit in the following situation. ・OS:Fedora23 ・Executable:etilgrmhd (IllinoisGRMHD) The compile was done and executable file was made. However the following error message was displayed when execute the file. "error while loading shared libraries: libgsl.so.19: cannot open shared object file: No such file or directory" There was no problem few weeks ago. So I don't have any idea why the error occured. What should I do? ・Update the OS? ・Update the IllinoisGRMHD? (I don't use the latest version) I would appreciate it if you would give me some advice. Thank you, Best regards.

2 1

minutes on ET workshop and school at NCSA discussion of 2017-05-01
by Roland Haas 01 May '17

01 May '17

Hello all, Present: (at least) Pablo, Frank, Gabrielle, Eloisa, Roland, Ian, Qian, Manuela, Erik Dates: 5 days during either the week of July 24th or week of July 31st Poll: please indicate your preferred dates by checking at least 5 consecutive days in in this doodle poll (Mo starting times will be prioritized, you can use ifneedsbe for secondary choices): http://doodle.com/poll/p7deb9npmgwwfbmv Please respond no later than *Thursday* as well will finalize room bookings for the lecture halls on Friday. If you would like to serve as an instructor/TA please let us (Roland Haas, Eliu Huerta, Gabrielle Allen or just list) know as well. Summary ------- Dates; 5 days of school (3 days) and developer workshop (2 days), during weeks of July 24th or July 31st Location: NCSA/UIUC, Urbana, IL, US target audience: students interested in numerical relativity in general and students working in groups that use the ET topics: * keep it hands on, provide theory enough to understand what the ET implements, eg 3+1 decomposition but not full derivation of BSSNOK formalism, similar for hydrodynamics * aim to teach students how to run the ET, make simple additions and modifications to existing code or add simple new thorns, stay clear of scheduling and AMR related details if possible, unless asked for by students * possibly have more formal lecture in the morning with more lectures interspersed with handson exercises in the afternoon, late morning. Detailed breakup into theory and hands-on still fluid and open for discussion. * include a visualization section of ~0.5 days * include data analysis of ET data in LIGO using the LIGO provided ipython notebooks even if only shown how to do this with sample data and including running the simulation https://losc.ligo.org/s/events/GW150914/GW150914_tutorial.html Detailed notes: * dates should be no earlier than July 24th due to restrictions and no later than 2nd week of August to avoid running into issues with students being in courses already * expect fewer students than in Trento (~45 students), mostly from Noth America, NCSA has room for about 50 participants we expect ~25 students * send questionnaire to students *before* school to gauge their level of expertise and what they hope to gain from attending the school. Use this to guide the level of detail of the course and exercises * pair up students so that knowledgeable and new users are paired up (this includes both science knowledge and also knowledge on eg command line tools and the ET) * ideas for topics: - see if track1 + track2 approach with more basic and more advanced exercises and topics is possible - consider intro into the ET and Cactus to (briefly) outline philosophy behind it - send or hand out questionnaire to students on last day to provide feedback on experience * provide some afternoon entertainment, eg tour of Blue Waters, organize dinner etc. * more hands-on approach on teaching likely works better with younger teachers that have more time to prepare, more formal lectures may work better with persons that have more experience teaching * aim to have extra TAs in classroom to help out students, aim for many TA (between 5-10 students per TA) 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://pgp.mit.edu .

1 0

meeting minutes for 2017-05-01
by Roland Haas 01 May '17

01 May '17

Present: Frank, Roberto, Steve, Roland, Eloisa, Gabrielle, Bhavesh, Erik, Manuela, Yosef, Zach, Qian, Eliu, Haris, Vassili, Pablo project presentation and reporting: * Gabrielle presented plans on more formal project reporting, https://hangouts.google.com/hangouts/redirect?dest=https%3A%2F%2Fdocs.googl… * would weed out ghost projects, developers and maintainers when projects no longer provide updates * allow to coordinate and organize projects "endorsed by the project" * "working groups" are longer lived entities, while "devlopment teams" are ephemeral that vanish once the project is done * action item: discuss offline and provide input, decide which working groups to participate in, will discuss again next week DataVault: * An Einstein Toolkit Data Repository https://drive.google.com/file/d/0B5-y-9jNuShAWVdUNmZHbm5pckk/view * Timeline summary: Year 1: prototype Develop components for generating metadata, basic archive services at NCSA. Year 2: development ...and add parameter files and a suite of initial data files to the DataVault. Year 3: user-ready Archive open for community data entry. Evaluation of tools. Year 4: tutorials to archive all of the data generated in the Einstein Toolkit DataVault * datasets will contain metadata describing it, some of it automatically generated some of it manually entered. Will contain doi URL for citation and reference purposes. * TBD: license of data, what metadate to include, where is the data stored will be sent out survey form to list next week * will also provide instructions on how to set up your own data storage eventually, however the primary goal is to explore the framework and find out what is required for a functional system * currently storage by third party repositories is not yet * currently datavault has funding for a small amount of storage, want to collect use cases and how to hook up datavault to institutional repositories. Currently sizes are limited to ~TB per year and GBs for individual files. * will discuss again next week after survey Project reports: NCSA: * see Qian's presentation RIT: * implement spherical coordinates BSSN code in ET codes in collaboration with Thomas Baumgarte and Zach Etienne. * working on infrastructure in Cactus to get boundary condition parallelized in ET (similar to rotatingsymmetry180 using slab thorn). Will be made available (incl. doc) once finished, currently being tested with a WaveMoL code which is working right now. * aim to have code available publicly in July * suggestion to make development code visible (at least to ET developers) so gather feedback eg on boundary condition * Zach and Thomas are working on generating code for the stencil in different coordinate systems. Will now merge the two sub-projects. * currently the arrays x,y,z contain r,theta,phi right now. Have to decide if we want to keep this or introduce a new "SphereGrid" thorn. Currently implementation was chosen due to ease of using existing code. * hydro equations are ready to be implemented, MHD equations are still being worked on * suggest spherical wave code as possible module in educational code repository that is part of SI2. More complex than Cartesian code so need to decide if extra complexity is acceptable. LSU: https://docs.google.com/document/d/1yX3rkcnsVbSovsh8_sZaqmZ8XEgGkQzggZ-EX09… * Steve Brandt has been working on a pre-sync branch of Carpet that automates synchronization and handles it in a lazy manner. This would eliminate one of the most tricky pieces in designing a Cactus schedule. A prototype exists on bitbucket https://bitbucket.org/cactuscode/cactus/branch/presync GT: * https://drive.google.com/file/d/0B3bzSynUjI9pQ3lnSURNbkpiNzQ/view * ID solutions mimicking astrophysical systems, currently single spherical system with different EOS * idea presented in: https://arxiv.org/pdf/1606.04881.pdf * infrastructure for binary neutron star simulations action items for future: * continue DataVault discussion after survey results (Qian) * provide presentation of math behind spherical coordinates code (Vassili) ET meeting in Europe: * maybe combine with meeting in Barcelona 16-20th http://astro-gr.org/astro-gr-barcelona-2017/ * mostly a retreat style for developers only ET meeting at NCSA: * want to have developer meeting in addition to retreat in Europe? * at least for the EU there is enough interest to justify two meeting * similar in North America * more details in separate mail to list 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://pgp.mit.edu .

1 1

meeting minutes for 2017-04-24
by Roland Haas 28 Apr '17

28 Apr '17

Present: Frank, Roland, Bhavesh, Ian, Steve, Roberto, Vassili, Eliu, Qian Zhang, Eliu Huerta KNL on stampede: * no update yet Include double-parsing in piraha for one release: * depends a on speed impact, expect to not be a lot but want to check * add text to error message on how to disable piraha * Roland to test speed of CST stage on Blue Waters ML_Constraints: * https://trac.einsteintoolkit.org/ticket/1995 * don't know if it "only" the constraints or also the evolution variables Datavault simulation data repository for ET data: * provide a central repository for simulation data * more complex search and curation interface than current databases (eg RIT http://ccrg.rit.edu/~RITCatalog/, GT http://www.einstein.gatech.edu/catalog/, SXS https://www.black-holes.org/waveforms/) * goal is to include a doi for the data and allow tracing of usage of data * would be more similar to yt hub: https://girder.hub.yt/#collections and whole tale https://opensource.ncsa.illinois.edu/projects/WT * would include the full set of files to reproduce the simulation * aims to raise awareness of data within LIGO * query for what type of metadata people have * for hydro simulations the data can be much larger than for vacuum since for hydro simulations there is more "interesting" data. Waveforms would be of comparable size * this could also be connected to the national data service (NDS) project at NCSA * will give a presentation next week Next ET release: * shift into June due to time costraints ET workshop * try for NCSA * try for July, to avoid running into teaching issues in August * want to sort this out as soon as possible to give people time to travel * to make it worthwhile to attend from Europe it would have to be multiple days * if school: decide on topics and announce very soon 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://pgp.mit.edu .

2 1

Re: [Users] [ET Trac] [Einstein Toolkit] #2031: Option to disable Piraha
by Erik Schnetter 24 Apr '17

24 Apr '17

I think the option should instead be to disable the old parser. The new parser already helped us find bugs in ccl files, and the new parser is the future. There is no known instance where the new parser silently have wrong results. Disabling it is being too cautious. -erik On Monday, April 24, 2017, Einstein Toolkit < trac-noreply(a)einsteintoolkit.org> wrote: > #2031: Option to disable Piraha > ----------------------+----------------------------------------------------- > Reporter: sbrandt | Owner: sbrandt > Type: defect | Status: assigned > Priority: unset | Milestone: > Component: Other | Version: development version > Resolution: | Keywords: > ----------------------+----------------------------------------------------- > Changes (by sbrandt): > > * owner: => sbrandt > * status: new => assigned > > > -- > Ticket URL: <https://trac.einsteintoolkit.org/ticket/2031#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)gmail.com> http://www.perimeterinstitute.ca/personal/eschnetter/

2 1

Einstein Toolkit Meeting Reminder
by Roland Haas 24 Apr '17

24 Apr '17

2 2

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

Users