LHCb Core Software Meeting

Date and Location

05. July 2006
10:00 - 11:20
CERN (2-R-030)


Florence, Hubert, Joel, Juan, Marco Cl. (minutes), Markus, Manuel, Nicolas, Olivier, Philippe, Thomas


Gloria, Marco Ca., Nick



Nothing to report

Software Releases

Gaudi (Hubert)

  • Preparing a CMT requirements file for GFAL (should work as a replacement for the common grid login script)
  • Switched to LCG 45, but something has changed and it is not working anymore.
    Florence said that that David Quarrie changed something, She will check.
Marco Ca.: It does not make sense to build Gaudi v18r6 against LCG_45, we need a working Root rfio client. Options are either to use LCG_44 (allows us to test Gaudi changes) or wait for LCG_45a. I have no preference.

LHCb, Boole, Brunel (Marco Cl. for Marco Ca.)

  • Brunel v30r5, for DC06 computing challenge rDST production is ready in my area.
    • Possible problem in ProtoParticle creation, waiting for Chris to check (he's due back today I think)
    • Huge improvement in TransportSvc speed (x5 in velo fit, x2 in forward fit) - see Juan for details
    • Plan to release early next week. Will use Gaudi v18r6 if available, but will not explicitly wait for it
  • There will be one more release of Brunel in July, for DST production after stripping. Waiting for:
    • Code for reading ETC in Brunel (being packaged by Juan/Patrick)
    • Code to selectively clean states from tracks (Matt working)
  • lxbuild020 has been reinstalled with slc4_ia32
  • Nick has set up a task force (leaded by Ulrik) to study streaming of real data. Mandate, membership etc. can be found at StreamingTaskForce.

Gauss (Florence for Gloria)

  • Gauss v25r0 based on Gaudi v18r5 is ready, but it is not tagged and Florance cannot release it.

DaVinci (Juan)

  • Trying to compare particles by ParticleMaker in DC06 (pruduced $ J/\psi $) and DC04 (does not work yet)
  • Slightly modified LAssembly. As a side effect, the trasport service is much faster (see the post on lhcb-reconstruction). Unfortunately it seems that the beam pipe is not seen anymore (Thomas). To be investigated.

Joel asked for the DaVinci version to use for stripping. Philippe suggested not to wait for final physics algorithms for the computing exercise. Juan will release v16r2 (as it is) based on Gaudi v18r5.

Panoramix (Thomas for Guy)

  • Trying to fix all the missing converters

Thomas noticed that there are 3 vertex classes: Vertex, PrimaryVertex and TrackVertex. The first two are known, but the role of the third one is not clear. (See also a recent thread on lhcb-reconstruction)

Round Table

Marco Cl.
  • As discussed last week, preared a script to set up the environment to use external libraries (as it is defined in LCGCMT). The script is using the version of LCGCMT specified on the command line, or in CMTPATH or used by the latest Gaudi. An email will be sent to announce it.
    Juan asked if it is possible to use a defined version of DaVinci instead. It is possible, in addition to the current default.
  • The next meeting, we will have a presentation on HyperNews by Tim Adye.
Complained about the absence of a "common" way of accessing Oracle DBs from Python. It should be enough to have cx_Oracle built for win32 too. (As of today, 11 Jul 2006, it is available. MCl)
  • Created and tested the workflow for DC06. To check with Marco Ca. if outputs are correct.
  • Try to prepare DaVinci+Brunel before vacation.

-- MarcoClemencic - 04 Jul 2006

Edit | Attach | Watch | Print version | History: r6 < r5 < r4 < r3 < r2 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r6 - 2007-11-07 - TWikiGuest
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    LHCb All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2020 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback