Meeting minutes:
Present were: Tanja, Roland, Frank, Erik, Bruno, Josh, Gabrielle, Ian, Christian, Yosef
* simfactory: ** the python based version of simfactory is ready for testing ** python based version will offer support for archiving, profiling and debugging beyond what is currently available in the perl version. ** Gabe suggested adding data management functionality ** a desire to make using simfactory easier was voiced, but no agreement on how to achieve this was reached
* reports on changes to WeylScal4 and the JPEG thorn(s)
* Erik and Gabe recapped the current status of the test accounts: ** they are full accounts on QueenBee sponsored by Erik ** their SUs come out of the Cactus allocation ** currently the tutorial is geared towards them and using QueenBee ** we would like a list of the test accounts (maybe showing which ones are active and which ones are dormant) ** wanted to follow up to see if there are users who never run a job but only compile the code
* it was suggested to use the Mercurial version of Carpet in the ET, however Erik says it is not yet stable and still contains serious bugs ** we need verification tests ** in particular for prolongation, restriction under realistic circumstances
* new thorns: ** Ian made his SystemStatistics thorn public (in AEITHorns) ** WaveToy will be included in the ET (RH: which version? The standalone ones or the MoL based one? There are a number of testsuites that use the later one.) ** TestPar a thorn to test parameter file parsing will be included ** EOS_Omni will be included in the source but stay deactivated.
* XiRel/CIGR anual report ** AMR: we are working on this, mostly on refluxing (Erik and Christian) ** MHD nearly there, working version for shocktubes, will start testing within a week, also test divB=0 constraint. currently work is concentrated on Con2Prim routines. (Josh and Scott, see wiki of the Carpet workshop: http://ccrg.rit.edu/~carpet/wiki/Main_Page) ** there has been no progress yet for higher order numerical methods ** there has been no progress yet including radiation or mircophysics into the code
* Yosef will give a technical talk about the numerical challenges when simulating extreme mass ratio inspirals (today)
Yours, Roland
- Erik and Gabe recapped the current status of the test accounts:
** they are full accounts on QueenBee sponsored by Erik ** their SUs come out of the Cactus allocation ** currently the tutorial is geared towards them and using QueenBee ** we would like a list of the test accounts (maybe showing which ones are active and which ones are dormant) ** wanted to follow up to see if there are users who never run a job but only compile the code
As an example of a user who has both compiled and run the code, I wonder if the instructions could be updated to include more detailed installation instructions for personal machines (e.g., one's own personal 'redshift'). I have run into intermittent dependency stumbling blocks. Being a prototypical test user who tries this in their spare time, any guidance in ironing out these issues -- particularly for a standard platform (osx) where a prototype already exists -- would be much appreciated.
On Mon, Sep 20, 2010 at 7:53 AM, Richard O'Shaughnessy oshaughn@gravity.phys.uwm.edu wrote:
- Erik and Gabe recapped the current status of the test accounts:
** they are full accounts on QueenBee sponsored by Erik ** their SUs come out of the Cactus allocation ** currently the tutorial is geared towards them and using QueenBee ** we would like a list of the test accounts (maybe showing which ones are active and which ones are dormant) ** wanted to follow up to see if there are users who never run a job but only compile the code
As an example of a user who has both compiled and run the code, I wonder if the instructions could be updated to include more detailed installation instructions for personal machines (e.g., one's own personal 'redshift'). I have run into intermittent dependency stumbling blocks. Being a prototypical test user who tries this in their spare time, any guidance in ironing out these issues -- particularly for a standard platform (osx) where a prototype already exists -- would be much appreciated.
The problem is usually not how to add the configuration options to an option list (be it Cactus or SimFactory), the problem is finding out what configuration options actually work. Cactus requires certain additional libraries not found on a standard Mac OS system, and there are many ways to install these.
We could (or should) describe one standard way to install these additional libraries, e.g. via the MacPorts system. I should really do this for Mac OS, and others should do this for Red Hat etc...
-erik
On 23/09/2010 03:07, Erik Schnetter wrote:
We could (or should) describe one standard way to install these additional libraries, e.g. via the MacPorts system. I should really do this for Mac OS, and others should do this for Red Hat etc...
I did this a while ago for Mac OS X with MacPorts when I was getting started with Cactus. It may need slight updating, but the attached file may be useful as a starting point.
Regards Barry
Barry
These are very good instructions. Do would want to create a wiki page for them on the Einstein Toolkit wiki? We can then begin to update them.
-erik
On Thu, Sep 23, 2010 at 10:26 AM, Barry Wardell barry.wardell@aei.mpg.de wrote:
On 23/09/2010 03:07, Erik Schnetter wrote:
We could (or should) describe one standard way to install these additional libraries, e.g. via the MacPorts system. I should really do this for Mac OS, and others should do this for Red Hat etc...
I did this a while ago for Mac OS X with MacPorts when I was getting started with Cactus. It may need slight updating, but the attached file may be useful as a starting point.
Regards Barry
Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users
OK, I've put them up here: http://docs.einsteintoolkit.org/et-docs/ConfiguringMacOSX
Barry
On 23/09/2010 17:33, Erik Schnetter wrote:
Barry
These are very good instructions. Do would want to create a wiki page for them on the Einstein Toolkit wiki? We can then begin to update them.
-erik
On Thu, Sep 23, 2010 at 10:26 AM, Barry Wardell barry.wardell@aei.mpg.de wrote:
On 23/09/2010 03:07, Erik Schnetter wrote:
We could (or should) describe one standard way to install these additional libraries, e.g. via the MacPorts system. I should really do this for Mac OS, and others should do this for Red Hat etc...
I did this a while ago for Mac OS X with MacPorts when I was getting started with Cactus. It may need slight updating, but the attached file may be useful as a starting point.
Regards Barry
Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users
Thanks.
I updated and rearranged it. I've taken out GSL and HDF5 to simplify things, since Cactus can now build these by itself.
-erik
On Thu, Sep 23, 2010 at 4:54 PM, Barry Wardell barry.wardell@aei.mpg.de wrote:
OK, I've put them up here: http://docs.einsteintoolkit.org/et-docs/ConfiguringMacOSX
Barry
On 23/09/2010 17:33, Erik Schnetter wrote:
Barry
These are very good instructions. Do would want to create a wiki page for them on the Einstein Toolkit wiki? We can then begin to update them.
-erik
On Thu, Sep 23, 2010 at 10:26 AM, Barry Wardell barry.wardell@aei.mpg.de wrote:
On 23/09/2010 03:07, Erik Schnetter wrote:
We could (or should) describe one standard way to install these additional libraries, e.g. via the MacPorts system. I should really do this for Mac OS, and others should do this for Red Hat etc...
I did this a while ago for Mac OS X with MacPorts when I was getting started with Cactus. It may need slight updating, but the attached file may be useful as a starting point.
Regards Barry
Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users
users@lists.einsteintoolkit.org
-
Barry Wardell -
Erik Schnetter -
Richard O'Shaughnessy -
Roland Haas