X-alignment workshop

The workshop was held at CERN on the 24th of May, 2011. The agenda can be found at the following address:

https://indico.cern.ch/conferenceDisplay.py?confId=140306

In the following some notes (more a summary of the issues than a complete report) that could serve as rough minutes/todo list.

Muon vs Tracker X-alignment

  • both the tracker alignment and the muon alignment expressed interest in using Sasha's machinery as a validation tool for each major update of the respective geometries since it can provide indications of relative movements of the muon and Tk geometries in the CMS reference frame. It would be extremely useful to have all the tools to run the workflow in the release and to setup some simple instructions on the twiki. Sasha, do you think this is feasible?

  • the workflow should possibly include muons crossing the CSC chambers instead of being limited to the barrel.

  • the possibility of introducing an ad-hoc Global-Position Record for the CSC detectors has also been discussed. Need to understand if the current implementation can be easily expanded and whether this is a real asset.

ECAL and ES alignment

  • the ECAl and ES alignment workflows are currently using mainly the W/Z RAW-RECO skims. While this was considered very convenient during the commissioning steps of the workflows there has been a general agreement that a dedicated ALCARECO dataset should be setup. The EcalElectron one is a potential candidate. We need to make sure that it contains all the products needed for the track refit and for the alignment both of the ECAL and of ES. We will follow up offline on the necessary steps to setup this stream.

  • ECAL and ES currently don't use the Global-Position Record. In principle this would allow to move the whole calorimeter+preshower ina coherent way w.r.t. all the other sub-detectors. It would be good if Andrea and Kai-Yi could have a look to how this is used and that it can be exploited in a consistent way within their workflows.

Beam-Spot calibration

  • Very often the co-existence of two tracker alignment geometries imposes the need of maintaining two parallel offline workflows for the beam-spot calibration. In principle, one could setup a simplified version of the workflow where the offset due to the different geometries are applied to all payloads produced with the "old/previous" tracker geometry so that there is no need to re-run on the full statistics. This would have minor impact on some of the quantities that need to be estimated. In principle the values of the offset in the central values of the position in the transverse plane could be directly extracted as a product of the tracker alignment procedure.

General considerations on the circulation of intermediate geometries for validation between various groups:

  • the usage of sqlite files is discouraged in favour of tag in the production DB, possibly with clear names and IOVs. This has clear advantages on the reproducibility of the results.

  • all groups (and in particular the muon alignment one) expressed interest in the circulation of "test/not final" geometries produced during the preliminary step of the tracker alignment since the X effects might have a large impact. The example of the twist-free geometry was brought up.

  • the creation of intermediate Global-Tags to be used by the various group during the finalisation of the whole X-alignments is considered useful.

  • the tracker experts are asked to advertise the configuration snippets used to refit the tracks instead of running the full reconstruction. In fact, refitting tracks is considered the right approach to apply new tracker geometries on existing reco/alcareco samples to be used for the study of the alignment of the various detectors.

DB tools considered useful:

  • tool for direct comparison of different payloads.
    cmscond_2XML can be used for this. I would invite everybody to try it out
  • tool to split a tag with several IOVs in several tags with IOV 1-inf. We are following this with the DB group, we asked to the DB guys to provide a centrally maintained tool based on the example provided by Andreas M.

  • dropbox: possibility to update simultaneously the prompt/express/hlt without necessarily uploading to an offline tag.

All this will be followed up with the DB group. In the meanwhile I invite everybody to have a look at: https://indico.cern.ch/getFile.py/access?contribId=9&sessionId=6&resId=1&materialId=slides&confId=88844

-- GianlucaCerminara - 06-Jun-2011

Edit | Attach | Watch | Print version | History: r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r1 - 2011-06-06 - GianlucaCerminara
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    CMSPublic All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2018 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback