Test/QA production 1e5 events
We make a request for a production to be used for second paper (physics and methodological studies) within our PWG3 in the scope of correlation femtoscopy.
UrQMD + Geant-3 based femtoscopy-purpose simulation with freeze-out coordinates project for minbias (b = 0-16 fm) Bi (83/209) +Bi (83/209) collisions at 9.2 GeV, detector configuration for femtoscopy only.
We request to produce a dataset with 100K events reconstructed for QA.
All necessary macroses:
runMC.C (11.7 KB)
runReco.C (6.4 KB)
Input root DST files at /eos/nica/mpd/users/gnigmat/mcDst/bibi_ecm_9.2
Thanks,
Konstantin Mikhaylov (konstantin.mikhaylov@cern.ch) on behalf of PWG3
Good afternoon.
Now there is a transition to a new version of Dirac, I can launch it in a couple of days. Resources of tier clusters are free, but available to me only through Dirac, resources available to me without Dirac now work with tasks of the 29th and 30th requests.
Hi, Andrey,
“in a couple of days” sound good to me!
Thanks,
Konstantin
I don’t have permission to copy from /eos/nica/mpd/users/gnigmat/mcDst/bibi_ecm_9.2/
Once the dirac is connected I will need to copy the input files into the dirac file catalog to work with them.
Now I can’t even look at the number of events in the files to make a production task template.
Dear Andrey,
I’ve informed Gregory about the problem.
I hope that Gregory will fixed the problem today.
Thanks,
Konstantin
Dear Andrey,
Just now I tested the directory. It seems to me everything is OK:
[kmikhail@ncx104 bibi_ecm_9.2]$ pwd
/eos/nica/mpd/users/gnigmat/mcDst/bibi_ecm_9.2
[kmikhail@ncx104 bibi_ecm_9.2]$ ls |wc
53760 53760 1935304
[kmikhail@ncx104 bibi_ecm_9.2]$ ls -l …/
total 0
drwxr-xr-x 1 gnigmat nica 0 авг 24 05:45 bibi_ecm_9.2
drwxr-xr-x 1 gnigmat nica 0 авг 20 10:10 bibi_kmikhail
Thanks,
Konstantin
Dear Konstantin,
There is a file access problem.
[moshkin@ncx101 cd /eos/nica/mpd/users/gnigmat/mcDst/bibi_ecm_9.2/
[moshkin@ncx101 bibi_ecm_9.2]$ ls -l
total 1756866827
-rw------- 1 gnigmat nica 18965710 Aug 16 16:48 urqmd_bibi_9.2gev_000000.mcDst.root
-rw------- 1 gnigmat nica 9928876 Aug 15 05:58 urqmd_bibi_9.2gev_000005.mcDst.root
-rw------- 1 gnigmat nica 10079757 Aug 15 19:25 urqmd_bibi_9.2gev_000009.mcDst.root
…
Hi Andrey,
I’ve updated the permissions which somehow, indeed, moved to the weird state. Please check the files again and let us know if everything is okay now.
Cheers,
Grigory
Hi Grigory,
Thank you, everything is OK. I hope today Igor will finish the transition of Dirac to the new version and will launch qa on tier clusters
QA took longer than expected, one task fell and got to the weakest node during recalculation, files of 1000 events are located in
/eos/nica/mpd/sim/data/qa/BiBi-09.2GeV-mp10-22-1kev-req31
1 Like
Dear Andrey,
Thanks a lot for such a quick production of 100K events in DST files.
I will look at them and let you know the results. It will take some times.
Best regards,
Konstantin
Dear @amoshkin ,
Could you please tell if any technical works are ongoing on the NCX cluster?
Over the last week or two it seems to be that EOS is systematically not mounted correctly
and files are not seen. Is it possible to fix or whom should we contact in order solve the issue?
Cheers,
Grigory
Good afternoon.
Cluster administrators have known the problems for a long time.
I don’t know when the problem will be solved, besides migrating to a new version of eos, there is also a problem with the failure of storage disks, the warranty period for which is already expiring. There is a copy of this data on the tier cluster in:
/eos/nica/mpd/dirac/mpd.nica.jinr/vo/mpd/data/exp/dst-BiBi-09.2GeV-mp10-22-1kev-req31/BiBi/09.2GeV-mb/PHQMD/BiBi-09.2GeV-mp10-22-1kev-req31
Interactive nodes of tier clusters are lxui0*.jinr.ru and lxpub0*.jinr.ru,
If necessary, I can make a copy on hybrilite.
Dear Andrey,
Could you write where are MiniDst.root files, please?
I need to look at primary tracks, but Dst file does not contain primary tracks.
Unfortunately global tracks only are in Dst.
Thanks,
Konstantin
minis on nica cluster in /eos/nica/mpd/sim/data/qa/req31/mini
Dear Andrey,
After some QA tests, I found a significant difference between request 25 and request 31. I would like to ask you to make a new 100 thousand events for request 31 with GEANT4.
As far as I understand, you need to change the type from EVMCType vmc = EVMCType::GEANT3 to EVMCType vmc = EVMCType::GEANT4
in runMC.C line:
void runMC(EGenerators generator = EGenerators::MCDST, EVMCType vmc = EVMCType::GEANT3, …
I understand that you may be busy right now before the New year. If so, then you can do it after the holidays.
Happy new Year and Merry Christmas!
Thanks a lot,
Konstantin
new QA
reco: /eos/nica/mpd/sim/data/qa/req31g4/dst/
MiniDst: /eos/nica/mpd/sim/data/qa/req31g4/mini/
Dear Andrey,
I found the problem of generating artificial events. I reported the details at the cross-pwd meeting January 31.
In this regard, I would like to ask you to make a new 100 thousand events for request 31 with the corrected file:
runMC.C (12.2 KB)
In this runMC.C, lines have been added to check the number of events in the input root DST files.
Please, let me know if you have questions.
Thanks a lot,
Konstantin
Done:
dst: /eos/nica/mpd/sim/data/qa/req31g4v2/dst/
mini:/ eos/nica/mpd/sim/data/qa/req31g4v2/mini/