I looked at the GRHydro test suite failures on Hopper (where we use a
PGI 9.x compiler). The failure is caused by differences that are
slightly larger than allowed, up to 2.5e-12 in absolute difference.
The differences occur in eps and in the metric.
What should we do? I am tempted to simply increase the tolerance there.
-erik
--
Erik Schnetter <schnetter(a)cct.lsu.edu> http://www.cct.lsu.edu/~eschnett/
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/info/support/
I short: the number is (+1) 225-578-4942 or (+1) 866-573-0359 and
the conference id is 118682#.
Some points of tomorrows discussion will be:
1) ET release plan for ET_2011_05
see https://trac.einsteintoolkit.org/wiki/ReleasePlan
2) ET paper plan:
-> April 04: Most text included, or sections get reassigned
April 11: All text included, data from example sections mostly there
April 18: Results from example sections included, or dropped
April 25: Text and data complete - "public review"
May 09: Addressed incoming comments, publish along with new release
Frank Loeffler
Hello all,
the servers hosting einsteintoolkit.org (website and a subversion) as
well as svn.cactuscode.org cannot be reached anymore (neither from GT
nor from eg. Ranger).
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.
Meeting minutes:
Present were: Frank, Tanja, Bruno, Roland, Christian, Eloisa, Peter,
Yosef, Ian, Erik
upcoming ET release:
* updated release timeline
https://trac.einsteintoolkit.org/wiki/ReleasePlan
* release will not include python simfactory or the Mercurial version of
Carpet, but we will switch to those about two weeks after the release
* major new feature of release: EOS_Omni
* will write transition guide to be either part of EOS_Omni
documentation or on the ET website (or both)
* MHD is still being tested and not yet ready for official release
** presentation of status at CIGR workshop
* testuites:
** remove obsolete machines: Abe, damiana
** add new machines: Datura (Ian Hinder), Lonestar and Hopper (Erik
Schnetter), Trestles (Christian Ott) incl. only if straightforward to
do, Pandora (Peter Diener)
** test with OpenMP (release with OMP only if no problems are found): 1
process+1 thread total, 1 process+2 threads total, 2 processes+2 threads
total
ET paper:
* publish independent of ET release (with timetable from Meeting Reminder):
April 04: Most text included, or sections get reassigned
April 11: All text included, data from example sections mostly there
April 18: Results from example sections included, or dropped
April 25: Text and data complete - "public review"
May 09: Addressed incoming comments, publish along with new release
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.