How to access PADME data - PADME-Experiment/padme-fw GitHub Wiki

To access the PADME data you must be logged on a GRID UI node (User Interface).

On the LNF Tier2 site the available UIs are atlasui, atlaswn170, atlaswn171, atlaswn180, all of them using CentOS 7. If you need an account on these UIs you can send an email to [email protected] to request it.

The padmeui server is also available but it is reserved for production activities: please contact Emanuele if you need to use it.

To access the data you must first create a VO-enabled proxy certificate based on your private certificate. The command to do this is:

voms-proxy-init --voms vo.padme.org

The system will ask for your PEM passphrase and will create a proxy certificate which is valid for 12 hours (you can extend it to 24 hours by adding the --valid 24:00 option to the command. 24 hours is the maximum validity time for a proxy) and can be used to access all PADME data, both at CNAF and at LNF.

As explained in PADME on the Grid, all PADME data are initially stored into the INFN-CNAF tape library where they can be accessed using the GFAL commands (documented here or here) at the URI srm://storm-fe-archive.cr.cnaf.infn.it:8444/srm/managerv2?SFN=/padmeTape

To, e.g., get a list of all runs acquired in 2020 the command to use is:

gfal-ls -l srm://storm-fe-archive.cr.cnaf.infn.it:8444/srm/managerv2?SFN=/padmeTape/daq/2020/rawdata

Sizeable portions of the data are also replicated to the PADME Tier2 site at INFN-LNF and can be accessed at the URI root://atlasse.lnf.infn.it//dpm/lnf.infn.it/home/vo.padme.org (please be aware of the double slash (/) in front of dpm in the path).

The local path to the files is identical at CNAF and at LNF, so to get a list of all runs acquired in 2020 which are currently available at LNF the command to use is:

gfal-ls -l root://atlasse.lnf.infn.it//dpm/lnf.infn.it/home/vo.padme.org/daq/2020/rawdata

Files can also be copied to a local disk area using the gfal-copy command. The copy can start either from the central repository at CNAF or, if available, from the LNF Tier2 storage system. Note that at CNAF files are usually stored on tape and require some time to be replicated to a local disk area before the actual copy of the file can start.

The command to copy a single file from CNAF to a local disk area (/tmp in the example) is:

gfal-copy srm://storm-fe-archive.cr.cnaf.infn.it:8444/srm/managerv2?SFN=/padmeTape/daq/2019/rawdata/run_0000000_20190201_161616/run_0000000_20190201_161616_lvl1_00_178.root file:///tmp/run_0000000_20190201_161616_lvl1_00_178.root

while the command for same operation but from the LNF Tier2 storage system is:

gfal-copy root://atlasse.lnf.infn.it//dpm/lnf.infn.it/home/vo.padme.org/daq/2019/rawdata/run_0000000_20190201_161616/run_0000000_20190201_161616_lvl1_00_178.root file:///tmp/run_0000000_20190201_161616_lvl1_00_178.root

To transfer a full run, one can use the TransferRun.py script available from the padme-cdr package.

If files are available at the LNF Tier2 site, they can also be directly accessed from PadmeReco and PadmeAnalysis using the xrootd protocol with URI root://atlasse.lnf.infn.it//dpm/lnf.infn.it/home/vo.padme.org and the usual local path.

As an example, the syntax to run PadmeReco on a single rawdata file is

./PadmeReco -i root://atlasse.lnf.infn.it//dpm/lnf.infn.it/home/vo.padme.org/daq/2020/rawdata/run_0030034_20200527_173023/run_0030034_20200527_173023_lvl1_04_062.root