Users

users@lists.einsteintoolkit.org

3 participants
3984 discussions

meeting minutes for 2016-09-26
by Roland Haas 30 Sep '16

30 Sep '16

Present: Frank, Matt, Roberto, Roland, Zach, Ian, Erik, Steve, Peter, Vassili Jenkins: * Ian and Barry are working on setting up the git super repository, hoping to get things up and running in a day or so Piraha: * Roland asks for higher level documentation, for example excerpts from the piraha paper Gallery: * consistency checks are done Hangouts: * Frank contacted google and it seems as if accounts for larger numbers of participants are institution accounts and are hard to get * Frank is investigating alternative VoIP setups, Roland mentions to use screenleap for slide presentations during talks (http://www.screenleap.com/) * LSU has a webex account that could be used, this is problematic for Linux users that have to call in via phone Hackathon topics: * include citiations in code. We had a discussion on how we want to include this, in the end we decided against it as there was not enough interest. * overlapping computation and communication using FunHPC. Erik has a proof of concept and it would be a starting point. https://bitbucket.org/eschnett/funhpc.cxx . We will try for this. * time will be decided on a doodle poll Docker container for the ET: * Ian has a container on dockerhub which can be found by searching for einsteintoolkit https://hub.docker.com/r/ianhinder/jenkins-einsteintoolkit/ 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 2

question about ML_BSSN::fixAdvectionTerms
by Slinker, Kyle Patrick 29 Sep '16

29 Sep '16

Hello, I'm wondering if anybody can tell me about the ML_BSSN::fixAdvectionTerms parameter (which I think was new in Somerville). I have been trying to reproduce some old results and discovered this morning that having this parameter set to "off" is what was causing the discrepancy. What does it do? The description "Modify driver and advection terms to work better?" is vague enough I don't know what to make of it. From McLachlan_BSSN.m it looks like it's switching how the advection terms in the 1+log and Gamma-driver conditions are handled. What changes is it making to these equations? And what is the purpose of them? If setting the parameter to "on" reproduces old results (and if it lives up to its name and fixes things), should "on" be the default behavior? Thanks. Kyle Slinker

3 3

Einstein Toolkit Meeting Reminder
by Roland Haas 25 Sep '16

25 Sep '16

Hello, 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 the URL: https://hangouts.google.com/call/5xt7voshalqmdcztr4zhfeox2ia Please suggest additional topics that you would like to add to the agenda. Roland Haas -- 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

EHfinder
by dlshcbmuipmam＠gmail.com 25 Sep '16

25 Sep '16

Hi, My aim is to find simulate the event horizon of binary blackhole. However, the documentation of EHfinder is hard to understand, and there is no existing parameter file on internet or inside the toolkit which use the EHfinder for my reference. (Well, there is a test par file inside ET, using exact Kerr metric, but it is not very useful) Can anyone introduce how to input the metric data into EHfinder? (I am not even sure should I do the EHfinder analysis with the simulation, or after the simulation then input the data in some ways) It is even better if you can provide an example par file using EHfinder. Thank you.

1 0

thorns providing citations
by Frank Loeffler 23 Sep '16

23 Sep '16

Hi, This is a proposal about how thorns can provide citations of resources (papers, ...), and how Cactus could use those to output a list of 'used resources'. I would like to make this the topic for the October-hack-a-day, if I find one or two more interested people. It shouldn't take more than that day to get something working. Before that, however, we should discuss what this should actually look like. The rest of this email is details about the proposal. If you are not interested, you can stop reading here. If you are, please let us know what you think after you've read it. While I said 'resources', I mean citations, usually to papers, but really it could be anything. There is also the question of how to cite something, and in the following I am assuming bibtex entries, as that is what most likely most people would find useful. First of all, thorns would need to have a way to provide citations. Thorns already provide all kinds of stuff, and most of that is 'announced to Cactus' in configuration.ccl, so let's use that, and not create yet another .ccl file: PROVIDES CITATION name AS "xyz" PROVIDES CITATION name IN bibtex filename, relative to thorn dir REQUIRES CITATION name This gives each citation a name, and would use that in a key-value table as key, where the value is a bibtex-entry, verbatim and (in this proposal) unparsed by Cactus. The name should ideally be the same as the key in the bibtex entry, but it doesn't need to be. We could decide to enforce equality, but then would have to parse the entry / bib file. This mechanism would only provide a way to define citations, and use them if you can be sure another thorn already provides it (REQUIRES). The next question would be how to tell Cactus that a certain citation should be included in the output, and what that output should look like. We already have two "classes" of citations in the ET: "requested" and "optional". "requested" citations should be cited whenever that piece of code was used, while "optional" are, well, optional. Thus, we should have a way to provide those. Also, whether a citation should be included might depend on parameters (we do have a few examples of those), and not just thorn activation. Thus, I believe the best way to include this is schedule.ccl: CITATION name CITATION OPTIONAL name These could be stand-alone, top-level statements, or be inside if-blocks, similar to, e.g., STORAGE statements. These statements only use the name of the citation as defined above - not the actual citation string itself. Once a citation is marked as 'being used', it will not be 'unmarked' within that run. Using this information, Cactus will then provide output regarding which citations are used, by which thorn, and of which type (requested, optional). This output is updated (overwritten) each time something is added (but waits until the regular thorn-activation sequence is completed). The output should, at least, consist of: - one bibtex file containing all known entries - two lists of keys, in machine readable format: requested/optional - machine readable information about which thorn requested what how Of course, depending on our wishes, part of this can and should be configurable via parameters. Frank Loeffler

1 0

running ML_ADMConstraints with timelevels=1 and no SYNCs
by Roland Haas 23 Sep '16

23 Sep '16

Hello all, I am wondering if there would be anything wrong with running ML_ADMConstraints with just a single time level for the constraints as well as using no SYNCs for it as long as I set it's calc_every parameter to the frequency of the coarsest level. Obviously this will give my "nonssense" values in the ghost zones, buffer zones and boundary (though on the boudary I can just set it to zero). However since the constraints are pure output and nothing ever depends on them the bad values only matter if they were to eg show up in the norms. It is my recollection that CarpetReduce actually does not use any values from the ghost zones (certainly that would be wrong) or buffer zones (this may be more recent). Is there any any other way the "bad" values could make it into the norms? Note that I do not worry about 1d, 2d, 3d output for the constraints. 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 3

use of NO_ALLOCATION in simfactory machine definition files
by Roland Haas 21 Sep '16

21 Sep '16

Hello all, I would like to remove the allocation = NO_ALLOCATION lines from simfactory's machine definition files. They seem to have not benefit that I can see and cause two types of problems: * simfactory does not default to the allocation in the [default] section which is the only one set up when doing sim setup on the cluster itself * simfactory does not warn/abort if no allocation is set for the cluster since as far as simfactory is concerned, there actually is an allocation defined (namely "NO_ALLOCATION"). This is eventually caught by qsub and friends but simfactory's handling of qsub errors is not very good unfortunately Does anyone remember why we introduced NO_ALLOCATION? And would anyone object to remove it from all the affected ini files (which is most of them it seems). 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

web site development
by Frank Loeffler 20 Sep '16

20 Sep '16

Hi, Our website needs an update. This can usually mean two things, and in our case it means both: content update/refresh and/or web technology update. This email and discussion should only be about the latter: which web technology to use. This email also got quite long. If you don't feel like reading through all this, please at least skip to the last paragraph and read that. What do we have now? Status quo is hand-written html with usage of php on every site, but mostly only for includes. There are exceptions, but there are only a few, and they can always be treated special. Why should we change the current setup? 1) The usage of php makes it necessary to use a local web server to view the pages before committing. While this isn't usually a big problem, because installing a small php-capable webserver is easy, it is a hurdle we rather shouldn't have when we want people to contribute to the pages. 2) The web layout doesn't look particularly nice on tablets, or even phones. While that was fine when the site was created, this should be addressed nowadays. What could we do instead? There might be more than these two issues (please speak up if you have more), but in the following, I will go only go into these two; separately because they don't really depend on each other in any way. 1) php We use php mainly to include common snippets, like headers and/or footers. php does this server-side, but because this requires a server to view the pages, we would have to use something else. There are only two possibilities. Either we do includes client-side, or we do them before the server sends out the pages: developer-side. Client-side the only available option is java-script, as that is the only 'active' component that is widely available client-side. Developer-side would mean that every web-developer would need to process includes after each change to the web pages, and would hand the result over to the web server. This can be automated, such that after each change to the repository a script (on the webserver) runs this step and creates the then static web pages. Requiring compilation on the developer-side would mean to require that every developer would need to install the tools needed for this, a situation we tried to avoid in the first place (local webserver). It would also mean to commit generated files into the web-repository, which in general is a bad idea. On the other hand, automating this step on the server would mean that the web sites couldn't be viewed locally again without installing the necessary tools. I would welcome something else than javascript that would work client-side, so we wouldn't depend on it being enabled. However, I am not aware of anything in that direction, so I see it as the best of the alternatives that we have. 2) layout The problem here is that it is difficult to create pages that look good both on large (desktop) and small screens (phones), especially if you think about plain html pages. The "usual" (as in often-used) solution is to change the layout of the page slightly depending on the size of the screen. Html doesn't allow for this, so again we arrive at javascript. There are probably tons of frameworks that implement common elements for so-called "responsive" web pages. Don't hesitate to throw in others, but I'll go on show-casing one I did play with. That one is called "Bootstrap". I am not saying it is the best, but it is one that seems to be widely used, open source, actively developed, and has tons of documentation. So I went ahead and used it to create a (not quite mock-up anymore) page for Chemora: https://chemoracode.cct.lsu.edu/ Note that this isn't mean to be a final "product", but it shows the basic feature of a menu. That menu is a simple top-navigation bar if inside a window that is wide enough to show it. If the windows size gets too small (you can try that yourself by just resizing your desktop window), this bar gets replaced by a pop-up menu. Feel free to look at the source code. You won't find a lot of scripting inside the actual pages. Most of that is hidden in files that are included, and most of that is inside the bootstrap library itself (that you don't really need to look at at all). How to get involved? I will have a student at LSU go through the pages and "improve" them, creating a "preview" first for everyone to look at. However, apart from a more "modern" page, one prime objective is to make it really, really easy to edit those pages, for every developer. Please share your opinion about how to edit/compose our web pages. How would *you* like to edit and view them? Frank

2 3

meeting minutes for 2016-09-19
by rhaas＠illinois.edu 20 Sep '16

20 Sep '16

Present: Jonah, Roberto, Roland, Eloisa, Matt, Zach, Erik, Frank, Vasili Mewes, Yosef * Jenkins move * will move to NCSA, master server up and running but requires backups to be still set up. git super repo needs to be moved. Roland to get this up and running. * Request for (final) review of https://trac.einsteintoolkit.org/ticket/1523 * Roland asks for documenation to be added to the Cactus maintainers guide, eg the grid paper Grid 2010 : CBHPC 2010 Workshop --- Session 3 * technical review should be done again * currently Piraha compares its parsing result with the old code so it is "safe" * Frank will apply for google academic institution to increase the number of participants to 15 * recap of the SI2 NSF grant kickoff meeting in Atlanta * Frank gave a quick overview * will send out the grant proposal to the mailing list and postpone until next week, will include sub-project list * Topics for monthly hack-a-days * (summer) student projects on https://docs.einsteintoolkit.org/et-docs/Summer_student_projects * ELoisa suggests to try and apply for a google summer of code project for some of the projects * An automated way to collect citation info from thorns, to be output at runtime. * postponed to next week, please contact knarf(a)cct.lsu.edu for suggestions * Decision on technical solution for web pages. * Bootstrap * remaining question: "offline" (jekyll) vs. "online" (javascript). * undergraduate student at LSU tasked with redoing the websites, currently learning the ropes. * Frank suggests using bootstrap for layout (can be used with Jekyll) * Erik suggests to create a couple of mock up pages using the different packages to see what itt takes and what it looks like . Currently we have one for Chemora: https://chemoracode.cct.lsu.edu/ * will move this to the mailing list * Ian suggest also include a content management system in the 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://keys.gnupg.net.

2 1

Einstein Toolkit Meeting Reminder
by Roland Haas 18 Sep '16

18 Sep '16

Hello, 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 the URL: https://hangouts.google.com/call/5xt7voshalqmdcztr4zhfeox2ia Topics: * recap of the SI2 NSF grant kickoff meeting in Atlanta * (summer) student projects on https://docs.einsteintoolkit.org/et-docs/Summer_student_projects Please suggest additional topics that you would like to add to the agenda. Roland Haas -- 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 1

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

Users