Present were: Frank, Roland, Barry Wardell, Erik, Ian, Steve, Bruno,
Peter, Josh, Philipp, Tanja, Christian
LSU server certificate:
* CCT has obtained a new certificate, https should now work
ET release:
* still open tickets [1]
* CZ4 formalism was added to McLachlan script. It generates an extra
thorn since it was found to negatively impact the current BSSN performance.
* Description will be added to McLachlan documentation
* include three wavetoy thorns: standard cactus, McLachlan using Kranc
and McLachlan using Kranc and OpenCL
* the last will be commented out by default with instructions how to
enable it in the thorn list.
* suggestion to have syntax in thorn lists to selectively enable thorns
but no working mechanism could be proposed
* will deprecate old library mechanism (ie HDF5=yes) with the exception
of MPI which used by many thorns without REQUIRE statements
* timing:
- soft freeze date of May 14th
- no additions (thorns ect.), major changes, new features
- hard freeze date of May 21th
- only 'important' bug fixes or documentation improvements
^--> to be decided by two maintainers
- release date of May 28th.
* include EinsteinExact arrangement by Barry with some documentation.
Bruno and Peter volunteer to look through code.
* migrate test to use new arrangement where possible for the next release
MHD paper:
* want to use a single plotting package
* supermongo disliked by those that do not have a license
* matplotlib disliked by current supermongo users
* provide matplotlib style files and samples
* encourage new plots in matplotlib which are easier to update to common
style than converting supermongo to matplotlib (or vice versa)
* spend day near end of paper writing process to convert all plots to
common style
* most tests that should work are done (ie. Balsara tests work,
Komissarow do not)
* TOV test works with constraint transport, fails in long runs with
divergence cleaning
* magnetic rotor test works with constraint transport, the LLF flux or
using +/-c for the characteristic speed for the longitudinal field component
* Bruno will do some characteristic structure analysis for proper speed
of longitudinal field component
Yours,
Roland
[1]
https://trac.einsteintoolkit.org/query?status=accepted&status=assigned&stat…
--
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.
_______________________________________________
Users mailing list
Users(a)einsteintoolkit.org
http://lists.einsteintoolkit.org/mailman/listinfo/users
On 05/07/2012 12:06 PM, knarf(a)cct.lsu.edu wrote:
> +# most machines don't support OpenMP yet,
Is this OpenMP or OpenCL? Just to make sure it isn't a typo.
CHeers,
B.
Hi
Please consider joining the weekly Einstein Toolkit phone call at
10 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/
I short: the number is (+1) 225-578-4942 or (+1) 866-573-0359 and
the conference id is 118682#.
The main topic will be the organization of the next ET release. In
particular, this means we will discuss:
- Release-relevant tickets (11) [1]
- CCZ4 formalism
- depreciate old library mechanism within Cactus
- Schedule:
- soft freeze date of May 14th
- no additions (thorns ect.), major changes, new features
- hard freeze date of May 21th
- only 'important' bug fixes or documentation improvements
^--> to be decided by two maintainers
- release date of May 28th.
Not necessarily connected to the release:
- Bug reports: quite a lot happened during last week: 40 issue were opened or
changed and are still open, and 28 tickets could be closed. Some of them
were:
- The ET trac has now an "reviewed ok" state for tickets with
solutions which have been reviewed, but are not yet committed.
- Problems with LSUs svn server (LSUThorns) have been resolved
- several external libraries received problem fixes
- The "Conformal Covariant Z4" formulation was implemented in
McLachlan, but the new thorns have not been made part of the ET - to
be discussed.
- Cactus requirements are now handled recursively.
- Several BSSN_MoL tests got updated for McLachlan
- Several simfactory host profiles received updates
- AEILocalInterp is now parallelized via OpenMP
After this, we received a request to discuss the current status of the
ET MHD paper, so after everything else has been discussed people
interested in this topic can just stay on the call.
As usual, feel free to add to this agenda.
Frank Loeffler
[1]
https://trac.einsteintoolkit.org/query?status=accepted&status=assigned&stat…
Hello all,
reading the docs at
http://simfactory.org/info/documentation/userguide/_auto/commands/run.html#…
it seems as if the "run" subcommand requires a --recover option to
recover from a checkpoint. However looking at the simfactory generated
RunScripts, this option is not used (though my runs seem to recover fine).
Since I have to do some lower-level fooling around with simfactory runs:
is this option still required? Or does simfactory by now find
checkpoints automatically and links them into the new folder?
In particular I am wondering if I can use a single qsub script
--8<--
#!/bins/h
# PBS ...
# PBS ...
sim run mysim --walltime 24:00:00 --procs 42 --num-threads 13
sim cleanup mysim
qsub $myself
--8<--
and have it re-submit itself again and again and have simfactory
properly create new output folders, link the checkpoints files from
segment to segment and run with the options.
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.