EUTelescope And ITK Testbeam pages

eutel-strategy.png


Glasgow work on ATLAS Pixel Upgrade Reconstruction and Analysis.

NB This page is a work in progress

Some pages

  • Information on finding data is given below.
  • EUTelescope pages are link ot EUTelescope
  • EUDAQ pages are also available here
  • UK activities are documented on Kate's pages on PUUKA
  • Find the working group meetings on indico

Instruction for customising reconstruction and analysis are also available

  • Some tips on post-analysis analysis are also noted.
  • Additional instruction for simulating telescope data with AllPix here

Link to the DESY March 2012 analysis (from 2013!).

Testbeam Reconstruction

There are five basic steps to run the reconstruction chain (i.e. to transform raw data into analysis ready data):

  • converter: Read in telescope and DUT hit information, from native DAQ to LCIO format
  • clustering: Combine individual hits on each plane to hit clusters on each plane
  • hitmaker: Globally orient planes
  • align: Global alignment of planes: based on pattern recognition & track fitting
  • fitter: Generate tracks -> device resolution

Each task is performed using the job sub command (exported when the build environment is set) and requires a configuration file to set basic steering parameters such as input file, output directories and beam energy. The basic example is $EUTELESCOPE/jobsub/examples/datura-noDUT.

For a more complete explanation see: ReconstructionDetails. Further official information on the software is available http://eutelescope.web.cern.ch/content/job-submission To install your own code, follow the documented instructions here.

Simple Recipe

This is a short recipe for running, based on ~March 2015 version of code. After installing and sourcing environment main directory is $EUTELESCOPE (equivalent to /PATH_TO_SOFTWARE/VERSION/Eutelescope/trunk).

  • First, check the environment has been set correctly by echoing the EUTELESCOPE parameter: 
>echo $EUTELESCOPE

This should be something like /PATH_TO_SOFTWARE/VERSION/Eutelescope/trunk. If not, then source $EUTELESCOPE/build_env.sh

  • Next, go to the directory you wish to work in, e.g. jobsub/examples/MYDIR:
> cd $EUTELESCOPE/jobsub/example/MYDIR

The basic execution format is > jobsub -c CONFIGURATION_FILE [-o PARAMETER_NAME=VALUE ] TASK_NAME FILE_NUMBER

  • Now you can proceed to run the reconstruction. Here is an example reconstruction chain (for datura-noDUT example):

> jobsub -c config.cfg -csv runlist.csv converter 97
> jobsub -c config.cfg -csv runlist.csv clustering 97
> jobsub -c config.cfg -csv runlist.csv hitmaker 97
> jobsub -c config.cfg -csv runlist.csv align 97
> jobsub -c config.cfg -csv runlist.csv fitter 97 The output of each step should include a root file, in $MY_OUTPUT/histograms, and an lcio file, in $MY_OUTPUT/lcio.

Link to CustomReconstruction.

Testbeam Analysis

The testbeam analysis software is called tbmon and can be found /data/detdev01/FEI4_testbeam_construction/analysis/tbmon and the tbmon source code is available on svn: svn address.

The software has three main configuration files: siteconfig.h, configs.cc and driver.cc, which are all found in the trunk directory. The siteconfig.h file is where default parameters such as run numbers and configuration settings can be defined, though these can be overwritten on the commandline. After any editting of the files the make command must be rerun.

NB Be sure the version of root you are using includes libraries for TFitResult and TFitResultPtr i.e. version >=5.26. If needed do

> source /data/ppe01/sl5x/x86_64/root/5.34.00/setup.sh

The basic execution format is

> ./tbmon -s/r/l RUN_NUMBER(s)/RUN_LIST -a ANALYSIS_TASK -c CONFIGURATION_SET

Other options can be added, for example -o OUTPUT_PATH and -d INPUT_PATH .

Here is an example analysis chain:

> cd PATH_TO_ANALYSIS/tbmon/trunk
> make
> ./tbmon -s 61527 -a hotpixelfinder -c eudetIBLsep2011
> cp OUTPUT_PATH/run61527-hotpixelfinder-*txt IBLcalibs/
edit driver.cc to set bool masknoisyanddeadpixels = true in relevant subroutine (e.g. here eudetIBLsep2011)

> make
> ./tbmon -s 61527 -a checkalign -c eudetIBLsep2011
> cp OUTPUT_PATH/run61527-checkalign-*txt IBLcalibs/
no need to edit or make here

> ./tbmon -s 61527 -a getetacorr -c eudetIBLsep2011
> cp OUTPUT_PATH/run61527-getetacorr-*txt IBLcalibs/
edit driver.cc to set bool applyetacorr = true and bool applytranslation = false

> make
> ./tbmon -s 61527 -a checkalign -c eudetIBLsep2011
edit driver.cc to set bool applytranslation = true and comment out hotpixelfinder, getetacorr and checkalign analysis objects from the allAnalyses function at the top of the file.

> make

for ease of use these steps are combined in the basic_analysis.sh file. In it the common tbmon command without a specified task (i.e. ./tbmon -s/r/l RUN_NUMBER(s)/RUN_LIST -c CONFIGURATION_SET) is set at the top to the basicCommand variable. This should be editted to refer to the correct and run set and siteconfig file. Also an uneditted driver.cc file should be copied to defaultdriver.cc to avoid any uneccessary file changes to the original driver.cc file. Once this is done the previous steps can be run in serial using the sole command > source basic_analysis.sh.

At this point any additional analyses can be run without further make commands. E.g. to perform residuals analysis:

> ./tbmon -s 61527 -a residuals -c eudetIBLsep2011
All standard analysis code resides in trunk/analysis/src. If no analysis is specified on the commandline then all analyses in the allAnalyses function of driver.cc are run.

The output of analyses are root files and pictures (in a format which can be specified in siteconfig.h or on the commandline with -e option).

For more details see some useful slides on tbmon commands and software.

Link to CustomAnalysis.

Tips on post-analysis analysis

To check the orientation of DUTs, use the clusterchecker_DUT_matchClusterSizeX and clusterchecker_DUT_matchClusterSizeX, with DUT appropriately substituted. One or other should relatively increase with any orthogonal tilt of the device. This can be compared to the recorded orientation of the device with respect to the beam.

Use maxcellres_DUT_chi2 and maxcellres_DUT_lv1 (with DUT appropriately substituted) to check the best chi2 and lv1 cuts in analysis.

Testbeam Data

Data can be found at Dortmund, DESY and CERN.

DESY

Test data can be run directly over AFS from DESY. The address is /afs/desy.de/group/telescopes/EutelTestData/.

Dortmund

Data at Dortmund must be copied locally and can only be accessed once an account has been set up on the Dortmund server. To do this you will need to contact andre.rummler@tu-dortmund.de .

Here is a spreadsheet on data sample and important parameters such as bias, theta and threshold.

Here is a link to ATLAS twiki. In the Testbeams there are links to the pages of each testbeam. Then on each page there is a link to a spreadsheet with run information hosted on google docs. You may need to ask for permission to view the data (just log-in to google, click to ask permission and wait for someone to add you, no questions asked).

CERN

Data can be found at CERN by first logging on to lxplus then sourcing the necessary EOS build: source /afs/cern.ch/project/eos/installation/atlas/etc/setup.sh

EOS can then be launched vai the eos command.

A list of testbeam data can be found: /eos/atlas/atlascerngroupdisk/pixel-upgrade/pps/BeamTest/ .

GRID

Requires a VO for ilc (done by DESY)

source /afs/cern.ch/project/gd/LCG-share/current/etc/profile.d/grid_env.sh

lcg-infosites --vo ilc lfc

export LFC_HOST=grid-lfc.desy.de

voms-proxy-init --voms ilc

Data can be found at:

lfc-ls -l /grid/ilc/aida-wp9

copy files (example)

lcg-cp --vo ilc lfn:/grid/ilc/aida-wp9/2012/tb-desy-spring/pps/joboutput/fitter/tbtrack999.root ./tbtrack999.root

-- KennyWraight - 2013-05-24

Topic attachments
I Attachment History Action Size Date Who Comment
Unknown file formatxlsx PPS_March2012_BatchSummary.xlsx r1 manage 90.4 K 2013-06-18 - 11:36 KennyWraight Spreadsheet with description of testbeam runs
Unknown file formatxlsx PPS_March2012_BatchSummary_extradetail.xlsx r1 manage 91.1 K 2013-06-27 - 15:38 KennyWraight Spreadsheet with description of testbeam runs
PDFpdf TBmon_Introduction_CERN_2012.pdf r1 manage 146.3 K 2013-05-29 - 09:46 KennyWraight Slides on tbmon analysis
PNGpng eutel-strategy.png r1 manage 106.6 K 2013-05-24 - 11:33 KennyWraight Schematic of EUTelescope strategy
PDFpdf howtoTBmon.pdf r1 manage 61.0 K 2013-05-29 - 10:00 KennyWraight full list of analysis options
PNGpng recon_chain.png r1 manage 58.4 K 2013-05-28 - 11:00 KennyWraight reconstruction chain flowchart
Topic revision: r30 - 2015-04-21 - KennyWraight
 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2008-2017 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback