Atlas Collaboration
TODO:
- user09
- LFC ACLs, script
- perf-jets, production role
- user.kaneda
- cleaning srmv1
- cleaning ccrc dark
Atlas Twiki
Atlas Computing
Atlas Computing Model
Event Data Model in ATLAS
http://doc.cern.ch//archive/cernrep/2005/2005-002/p255.pdf
The ATLAS Computing Model
ATL-SOFT-2004-007; ATL-COM-SOFT-2004-009; CERN-ATL-COM-SOFT-2004-009; CERN-LHCC-2004-037-G-085.- Geneva : CERN, 2005 - 37
http://documents.cern.ch/cgi-bin/setlink?base=atlnot&categ=Note&id=soft-2004-007
yellow report
http://doc.cern.ch//archive/cernrep/2005/2005-002/p1007.pdf
Atlas Computing Operations
Monitoring
Support
GROUPDISK
CERNDISK (atlascernuserdisk, atlascerngroupdisk)
AGIS
FroNTier / Squid
CVMFS
CERN/IT announcement
CS-OPS
Twitter
Savannah
Local Computing
LocalComputing (GridKa, Swiss (UNIGE, etc.), Spanish T2s)
Atlas Production
- AKTR (ATLAS Knowledge management of Task Requests)
Reprocessing
Group Production
Dataset Information
Meta Data
Streams
|
Correct |
Reality |
|
Tile |
00000001 |
00000101 |
RPC |
00000010 |
00000110 |
TGC |
00000100 |
00000110 |
CTP |
00001000 |
00000011 |
TAG
- 2008.11.05 Florbela Tique Aires Viegas slides
DDM
IUedaAtlasDDM
Nice plots
- dashb v2
- http://dashb-atlas-data.cern.ch/doc/guides/web-data/html/user/faqAppendix.html#id269911
- dst_plots http://dashb-atlas-data.cern.ch/dashboard/ddm2/#from_date=202401010000&to_date=202406010000&interval=0&p_height=570&p_label_step=7&p_bin=d&p_width=970&tab=dst_plots
- activity_plots http://dashb-atlas-data.cern.ch/dashboard/ddm2/#from_date=202401010000&to_date=202406010000&interval=0&p_height=570&p_width=970&p_bin=d&tab=activity_plots
- alpha http://dashb-atlas-data.cern.ch/dashboard/ddm2/#dst_site=%28UKI-LT2-QMUL_DATADSK,DESY-ZN_DATADSK,DESY-HH_DATADSK,INFN-ROMA1_DATADSK,INFN-MILANO-ATLASC_DATADSK,LRZ-LMU_DATADSK,IFAE_DATADSK,UKI-NORTHGRID-MAN-HEP_DATADSK,INFN-NAPOLI-ATLAS_DATADSK,MPPMU_DATADSK,IFIC-LCG2_DATADSK,MWT2_DATADSK,NET2_DATADSK,SLACXRD_DATADSK,SWT2_CPB_DATADSK,UKI-SCOTGRID-GLASGOW_DATADSK,GRIF-LPNHE_DATADSK%29&from_date=201101010000&interval=0&p_bin=d&p_height=570&p_label_step=7&p_width=970&tab=dst_plots&to_date=201108310000
- bravo http://dashb-atlas-data.cern.ch/dashboard/ddm2/#dst_site=%28IN2P3-CPPM_DATADSK,UKI-SOUTHGRID-BHAM-HEP_DATADSK,UKI-SOUTHGRID-RALPP_DATADSK,GRIF-IRFU_DATADSK,RU-PROTVINO-IHEP_DATADSK,
LRZ-LMU_DATADSK,IFAE_DATADSK,UKI-NORTHGRID-MAN-HEP_DATADSK,INFN-NAPOLI-ATLAS_DATADSK,MPPMU_DATADSK,IFIC-LCG2_DATADSK,MWT2_DATADSK,NET2_DATADSK,SLACXRD_DATADSK,SWT2_CPB_DATADSK,UKI-SCOTGRID-GLASGOW_DATADSK,GRIF-LPNHE_DATADSK%29&from_date=201101010000&interval=0&p_bin=d&p_height=570&p_label_step=7&p_width=970&tab=dst_plots&to_date=201108310000
Panda
Non-Pledged Resources
- Beyond-Pledge Resources Management ( Panda Server by MAENO, Tadashi on 02-Dec-2010 )
- Runs jobs for users who belong to a particular country group if those jobs are waiting in the queue. Otherwise, runs any user’s jobs
- availableCPU,pledgedCPU,availableStorage, pledgedStorage in schedconfig
- The pilot sends a request with countryGroup=XYZ and allowOtherCountry=True -- Single pilot stream
- The panda server does
- calculate the ratio of running jobs with countryGroup=XYZ to all running jobs at the site
- Get the number of activates jobs with countryGroup=XYZ
- A job with countryGroup=XYZ is sent to the pilot when the ratio is less than
(1 – pledgedCPU/availableCPU)
and there are some activated jobs with countryGroup=XYZ. Otherwise, any job is sent
- old way of Site-level access control with "Country group" and "Working group" ( pAthena/Panda developments by MAENO, Tadashi on 02-Sep-2009 )
- Country group
- Defined in VOMS
- Extracted from user’s proxy when jobs are submitted
- Working group
- More general list of members
- Can be defined in PandaDB arbitrarily for each site. ( PandaUserInfoTool )
- Users can set -–workingGroup option when they belong to the workingGroup at the site. The option is available since 0.1.65
Grid
- To use the grid certificate loaded on browser for accessing CERN protected pages
LCG-FR
Storage at CERN
Castor
Distributed Analysis
Ganga
Pathena
Database
Atlas Software
This is done by the deployment team (see
AtlasSoftOnLCG2 and also the talks by A.DS
comp-opr 15.dec.2006,
at SW week 12.sep.2006). The current deployment status can also be seen by following the link there.
In case the necessary atlas software release is not installed in a site,
a request should be made following the instructions there.
The releases currently used in productions;
SL5
Atlas Software
Atlas Simulation
IUedaAtlasSimulation
Atlas Data
Atlas Operation Tasks
TGC
WLCG
Python
- SubversionForADC
- https://svnweb.cern.ch/trac/lcgdm/wiki/CvsSvnMigration
- cvs to svn migration note from Sergey.Belov to vincent.garonne
- Dumping CVS repository:
> cp -a /net/dashboard.cvs.cern.ch/local/projects/dashboard/rep cvs
> afs/cern.ch/project/svn/cvs2svn/cvs2svn-2.1.1/cvs2svn --use-cvs
--fallback-encoding=ascii --dumpfile=lcg-dashb-dump --trunk=/trunk
--branches=/branches --tags=/tags cvs
- And restoring it to SVN:
svnadmin load /afs/cern.ch/project/svn/reps/dashboard < lcg-dashb-dump
- All modified scripts and configs (to support SVN instead of CVS) would be placed to the new repository.
$ svn merge -r 303:302 http://svn.example.com/repos/calc/trunk
U integer.c
$ svn status
M integer.c
$ svn diff
…
# verify that the change is removed
…
$ svn commit -m "Undoing change committed in r303."
Sending integer.c
Transmitting file data .
Committed revision 350.
Tips
LD_LIBRARY
>
lcg-cp: symbol lookup error:
>
/afs/cern.ch/project/gd/LCG-share/3.1.8-0/lcg/lib/liblcg_util.so:
>
undefined symbol: gfal_set_vo could not finish 'lcg-cp -v --vo atlas gsi
$ ldd `which lcg-cp` | grep gfal
which should give
libgfal.so => /afs/cern.ch/project/gd/LCG-share/3.1.8-0/lcg/lib/libgfal.so (0xf7b73000)
gfal_set_vo is definded in the so;
$ nm -C /afs/cern.ch/project/gd/LCG-share/3.1.8-0/lcg/lib/libgfal.so | grep gfal_set_vo
000000000001120f T gfal_set_vo
https://hypernews.cern.ch/HyperNews/Atlas/get/DDMOperations/380/1.html
Mac-Windows Image Incompatibility in PowerPoint
screen shot / screen dump / window dump / snap shot
ML
lxpuls
- Tue 29th June 2010: Users running i/o intensive work are requested to start your work with: e.g: ionice -c 2 -n 3 ./thrashTmp.sh
etc
InterWikis
ELOG:4822
GGUS:50570
BUG:53654
-- Main.iueda - 14 Jul 2006
- Set ALLOWTOPICCHANGE = UedaI
Topic revision: r383 - 2013-09-24
- UedaI