Meeting minutes:
Present were: Tanja, Roland, Bruno, Erik, Ian, Yosef, Manuella
Citation policy:
* Erik will send an email to the user mailing list and to the authors
mentioned in the README files of contributed thorns, asking for which
paper should.must be cited to use the thorn
* eventually we will collect this information in a form that can be
parsed in one of the thorn description files
MHD:
* the development copy at RIT has been committed to the repository
* done are HLLE, PPM, TVD, ENO
* still missing are more in-depth tests, divergence cleaning,
Con2Prim_polytype
Carpet Mercurial:
* please test refluxing
* error in reduction operation has been fixed
* it should now be ready for beta-testing
Einstein Toolkit release:
* will likely have the next ET release in the week "right after the
SEASAPS meeting (the week of October 25), and still before Supercomputing"
New modules:
* NoMPI: minimal set of MPI stub routines
* GetComponents moved to GitHub, we will need to decide whether to use
GitHub or distribute our own branch/copy in subversion
User support:
* Erik, Frank and Eloisa have started weekly Cactus support hotlines
using Evo, scheduled Wednesdays 9:00am Central Time (Evo meeting:
"Cactus"),
http://evo.caltech.edu/evoNext/koala.jnlp?meeting=eleBevvsvIa8avaeIl
We will have a presentation on simfactory during the next phone conference.
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.
This is a brief reminder about our regular EinsteinToolkit phone call
later today at 10:00 CDT. I apologise for the late notice.
Instructions for joining the call are at
<http://einsteintoolkit.org/info/support/>.
-erik
--
Erik Schnetter <schnetter(a)cct.lsu.edu> http://www.cct.lsu.edu/~eschnett/
Meeting minutes:
Present were: Tanja, ROland ,Frank, Gabrielle, Scoot, Joshua, Yosef,
Eloisa, Erik, Bruno
Carpet workshop recap:
* refluxing is implemented (for multiple processors) but needs more
testing in production sized runs
** it requires Carpet support as well as an external refluxing thorn
** refluxing adds constraints to how the grid is laid out that have to
be enforced by the regridding thorn
** CarpetRegrid2 adheres to these constraints
** it now works on multiple processors
* MHD is largely implemented, implemented are HLL, Con2Prim, Prim2COn,
PPM, still needed is Con2Prim_Polytype. It passes simple shocktube tests
but requires more testing.
* the unused grid points mask is implemented but it turns out it is only
non-empty if the grids are very large
* we know what to do to support higher numbers of refinement levels, but
implementing it requires carefully going through the code
* the code to not output 3d buffer zones is written but not well tested,
it will be committed to the Mercurial version
* I/O
** Nirvana (N5) library has been written and provides a high level
interface to I/O similar to Werner Bengers F5 format
** ADIOS 2.0 has been made public, and now supports AMR, so it might be
a useful backend for N5
** Christian Reiswig has added code to create subdirectories for the
output files to avoid having tens of thousands of files in a single
directory
* Ian Hinder has written a timer class and a Mathematica notebook to
visualize the output of the class
Citation policy:
* thorn authors are free to specify which citations they would like to
see, the Toolkit does not mandate any particular requirements (or lack of)
* files listing both the compulsory and recommended citations will be
part of the Einstein Toolkit checkout (possibly auto-generated form the
thorn sources)
We will have a presentation on simfactory during the next (or soon
after) phone conference.
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.
