AF
meeting 9th December
2010
Attendees
Marco
Clemencic, Marco Cattaneo, Pere Mato, Andrea Valassi, David
Rousseau, Pete Elmer, Rene Brun, Liz Sexton-Kennedy, John Harvey, John
Apostolakis, Predrag Buncic, Oliver Keeble, Fons Rademakers, Paolo
Calafiura [remote] , Sebastien [remote]
Line Management Issues
Action Items
Priorities and Organization of the PH-SFT group for 2011-2012
(Pere)
Experiments Feedback
AoB
version
1 23/12/2010 @ 14:10)
Line
Management Issues (Pere)
- Technical Student committee
meeting took place. There have been two technical students assigned to
the SFT group (one
of them has free choice between SFT and IT)
Action
Items
- Provide
test suites for dcache, xrootd, etc. in roottest (Rene)
- Include
the standalone test suites into the nightly builds (Stefan)
- Document
what python modules exist in the external area (Benedikt)
- Clarification
of the status of the '_cov' platform in
the nightlies (Benedikt)
Priorities and Organization of
the PH-SFT group for 2011-2012
Pere's
presentation can be found
attached in the agenda page. The main message is the proposal to
strengthen of role of the AF meetings. We would like to use the AF for
the strategic planning of the contents and the schedule of releases for
all the SFT products, by setting priorities and monitor progress of the
to-do list of each project. It should also be the channel for the
experiments to provide feedback and new requests for the services
provided by the SFT group. The current scope should be enlarged by
incorporating simulation and analysis use cases. This may need some
adjustments the experiment’s representation to cope with this
enlargement.
The new organization of the SFT group was well received. Concerning the
AF, the general impression was that it had already the proposed role,
so it was not very clear initially the need for any change. Therefore
it was very easy to agree with the continuation of the role of the AF
and also with the proposed minor changes as well as the scope
enlargements.
It was proposed also in the discussion to organize software seminars in
a regular manner to address a large audience on new developments and
success stories. This should also be a mechanism to receive feedback
from
direct users of the software developed and maintained by SFT.
Experiments
Feedback
- ATLAS
- There will be a major
release mid-January.
- There is no plans to
migrate to 64 bit.
- CMS
- Currently converging with
ROOT version 5.27/06. Fixing the last TMVA issues ...
- There is positive feedback
from Geant4 version 9.4 so far since it 'looks the same'.
- They plan to use 64 bit in
production. The main limitation was the online system, which is now
progressing well.
- LHCb
- Trying
to keep 32 bit support (needed for CernVM for 32 hosts).
- They
still need the slc4 binaries for some problem in the Geant4/Gauss
interface.
- They
would like to have a new LCGCMT configuration with the new ROOT
included. -->
Action Benedikt
- They
required to have ROOT working with the compiler gcc 3.4. This is needed
for PVSS systems integrating ROOT histograms.
- Ongoing
review of the HLT performance. Envisaging the idea to the have icc
compiler on the
HLT. They will be testing this option.
- Alice
- Preparing
for the ROOT release.
AoB
- The AF was invited to the
drink to celebrate the successful end of the year releases of Geant4,
ROOT, CernVM.