Skip to content

Latest commit

 

History

History
77 lines (58 loc) · 3.29 KB

Readme.md

File metadata and controls

77 lines (58 loc) · 3.29 KB
Raw

REANA example - FCC-hh Full Simulation Tracker Performance study

About

This example reproduces one curve of figure 7.12 from the Conceptual Design Report (CDR) of the hadronic Future Circular Collider (FCC-hh) http://cds.cern.ch/record/2651300 .

CDR figure 7.12

Fig. 7.12 compares the momentum resolution of the proposed track detector model for two magnet options. The resolution is calculated using different approximations. Here we reproduce the full simulation result that simulates single muons in Geant4 and fits them using the Riemann fit, all using the FCC software framework FCCSW.

The first step is the full detector simulation of single muons. This should take around 30 mins on a machine with at least 4GB of RAM. The output root file contains information on the original particles and the corresponding tracker hits in the FCC event data model. We can inspect the number of hits per track with the script 'numHitsPerTrack.C("muons_for_seeding_discrete_pt.root")', for example.

numHitsPerTrack

The final step is fitting the tracker hits with a Riemann fit, implemented in the tricktrack seeding library, which is again integrated in the FCC Framework. This yields the reconstructed tracks, which can be plotted with the python script plot_single_particle_resolutions.py:

finalResolutions

The pink curve corresponds to the one in figure 7.12. While the Riemann Fit cannot correctly estimate the resolutions for low momentum tracks, whose errors are dominated by scattering, the agreement of the high momentum tracks with the analytical methods is good.

How to run the example on REANA cloud

REANA is a reproducible and reusable research data analysis platform that permits to run structured computational workflows on remote compute clouds.

  1. First, set up reana-client that will be used to submit the analysis to REANA cloud:

    $ # on lxplus.cern.ch, you can use a provided environment:
$ source ~simko/public/reana/bin/activate

$ # on your laptop, you can create one as follows:
$ virtualenv ~/.virtualenvs/reana
$ source ~/.virtualenvs/reana/bin/activate
$ pip install reana-client  # FIXME: needs a new release on PyPI

  2. Second, configure the client to connect the wanted REANA instance where the analysis will run, using the access token that was provided to you by the REANA instance administrators:

    $ export REANA_SERVER_URL=http://reana.cern.ch/
$ export REANA_ACCESS_TOKEN=XXXXXXX

  3. Third, you can now use reana-client to create new workflow, upload input code, start the execution, check its status, logs, produced output files and download the plots:

    $ reana-client create -n fcchh-fullsim
$ export REANA_WORKON=fcchh-fullsim
$ reana-client upload
$ reana-client start
$ reana-client status
$ reana-client logs
$ reana-client list
$ reana-client download numHitsPerTrack.png

For more information on how to interact with the REANA platform, please see reana-client documentation.