The priorityuser role

Policy and Purpose

PAGs, POGs, and DPGs are allowed to specify a limited number of people who will have special privileges at the Tier-2s, both for CPU as well as write access to storage. The special privileges entail:

  • priority batch CPU access to half of the 50% of the pledged resources for analysis at all Tier-2s.
  • write access into /store/group/ at all Tier-2s associated to that group.

We assume that on average a lot less than 25% of the CPU resources at the Tier-2s will be used that way. The policy is thus set up for a burst but not average use. We will monitor usage.

The space used in /store/group/ will have to be accounted by the site's system adminstrators towards the quote the group has at that site.

Conveners' Responsibility

The conveners fill out the following table to indicate their priority users, and then send an email to the AnalysisOps hypernews for information,also mention who has been removed from the list. They do this every time they want to change something. However, we would like the groups to not change the names more often than once every few months. We will try to keep the actual privileges available as implemented in the CMS VOMS in synch with this Twiki table. It will be the responsibility of the people listed there to subscribe to this role using the CMS VOMRS interface. It should take not more than a few working days to get an approval email. The dynamic list of people authorized for the priorityuser role is here

  • user's DN is the output of the command: grid-proxy-info -identity (this will be used to access the StoreResults status web information, see below)

Phedex group name DN given name email address
AnalysisOps /DC=org/DC=doegrids/OU=People/CN=Frank Wuerthwein 699373 Frank Wuerthwein fkw at ucsd dot edu
AnalysisOps /O=GermanGrid/OU=RWTH/CN=Thomas Kress Thomas Kress thomas dot kress at cern dot ch
AnalysisOps /C=IT/O=INFN/OU=Personal Certificate/L=Trieste/CN=Stefano Belforte Stefano Belforte stefano dot belforte at ts dot infn dot it
AnalysisOps /DC=org/DC=doegrids/OU=People/CN=Sanjay Padhi 496075 Sanjay Padhi sanjay dot padhi at cern dot ch
AnalysisOps /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=dlevans/CN=638933 David Evans d dot evans at cern dot ch
AnalysisOps /O=GermanGrid/OU=RWTH/CN=Manuel Giffels Manuel Giffels manuel dot giffels at cern dot ch
tau /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=slehti/CN=408791/CN=Sami Lehti Sami Lehti sami dot lehti at cern dot ch
tau /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=rwolf/CN=667992/CN=Roger Wolf Roger Wolf roger dot wolf at cern dot ch
b-physics /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=griley/CN=731209/CN=Grant Riley Grant Riley gvriley at gmail dot com
b-physics /C=CN/O=HEP/O=PKU/OU=PHYS/CN=Linlin Zhang Linlin Zhang Linlin dot Zhang at cern dot ch
b-physics /DC=org/DC=doegrids/OU=People/CN=Elif Asli Albayrak 560229 Elif Asli Albayrak eaa at cern dot ch
exotica /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=srimanob/CN=628708/CN=Phat Srimanobhas Phat Srimanobhas srimanob at cern dot ch
ewk /C=IT/O=INFN/OU=Personal Certificate/L=Napoli/CN=francesco fabozzi Francesco Fabozzi francesco dot fabozzi at cern dot ch
ewk /C=GR/O=HellasGrid/OU=demokritos.gr/CN=Stilianos Kesisoglou Stelios Kesisoglou stilianos dot kesisoglou at cern dot ch
forward /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=mpetek/CN=676136/CN=Marko Petek marko petek marko dot petek at cern dot ch
top /C=DE/O=GermanGrid/OU=RWTH/CN=Felix Hoehle Felix Höhle felix dot hoehle at cern dot ch
top /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=gbenelli/CN=485942/CN=Gabriele Benelli Gabriele Benelli gabriele dot benelli at cern dot ch
tracker-pog /DC=org/DC=doegrids/OU=People/CN=Boris Mangano Boris Mangano boris dot mangano at cern dot ch
tracker-pog /DC=org/DC=doegrids/OU=People/CN=Brian R Drell 388439 Brian Drell brian dot drell at cern dot ch
muon /C=IT/O=INFN/OU=Personal Certificate/L=Bologna/CN=Alessandra Fanfani Alessandra Fanfani alessandra dot fanfani at cern dot ch
moun /C=IT/O=INFN/OU=Personal Certificate/L=Sns/CN=Giovanni Petrucciani Giovanni Petrucciani giovanni dot petrucciani at cern dot ch
muon /DC=org/DC=doegrids/OU=People/CN=Andrew Kubik 945511 Andrew Kubik andrew dot michael dot kubik at cern dot ch
susy /C=DE/O=GermanGrid/OU=uni-hamburg/CN=Lukas Vanelderen Lukas Vanelderen lukasvanelderen at gmail dot com
susy /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=alkaloge/CN=654885/CN=Alexis Kalogeropoulos Alexis Kalogeropoulos Alexis dot Kalogeropoulos at cern dot ch
higgs /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=jfernan/CN=529280/CN=Javier Fernandez Menendez Javier Fernandez Menendez javier dot fernandez at cern dot ch
higgs /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=klute/CN=529652/CN=Markus Klute Markus Klute markus dot klute at cern dot ch
qcd /DC=org/DC=doegrids/OU=People/CN=Vanessa Gaultney 303904 Vanessa Gaulney vanessa dot gaultney at cern dot ch
qcd /DC=org/DC=doegrids/OU=People/CN=Suvadeep Bose 616377 Suvadeep Bose suvadeep dot bose at cern dot ch
qcd /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=yuanchao/CN=596728/CN=Yuan Chao Yuan CHAO yuan dot chao at cern dot ch
jets-met_hcal /DC=org/DC=doegrids/OU=People/CN=Konstantinos Kousouris 2271 Konstantinos Kousouris konstantinos dot kousouris at cern dot ch
jets-met_hcal /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=delre/CN=651284/CN=Daniele Del Re Daniele del Re daniele dot del dot re at cern dot ch
e-gamma_ecal /C=UK/O=eScience/OU=Imperial/L=Physics/CN=nikolaos rompotis Nikolaos Rompotis nikolaos dot rompotis at cern dot ch
e-gamma_ecal /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=meridian/CN=555986/CN=Paolo Meridiani Paolo Meridiani paolo dot meridiani at cern dot ch  
b-tagging /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=adamw/CN=369199/CN=Wolfgang Adam Wolfgang Adam wolfgang dot adam at cern dot ch
b-tagging /DC=org/DC=doegrids/OU=People/CN=Pratima Jindal 522031 Pratima Jindal pratima dot jindal at cern dotch
generators /C=IT/O=INFN/OU=Personal Certificate/L=Trieste/CN=Fabio Cossutti Fabio Cossutti fabio dot cossutti at cern dot ch  
heavy-ions /DC=org/DC=doegrids/OU=People/CN=Wei Li 344273 Wei Li wei dot li at cern dot ch
B2G /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=bazterra/CN=667990/CN=Victor Eduardo Bazterra Victor Bazterra baites at fnal dot gov
B2G /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=ajkumar/CN=718734/CN=Ajay Kumar Ajay Kumar a dot kumar at cern dot ch
B2G /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=jdamgov/CN=537562/CN=Jordan Damgov Jordan Damgov jdamgov at fnal dot gov
B2G /DC=ch/DC=cern/OU=Organic Units/OU=Users/CN=jdolen/CN=674229/CN=James William Dolen James Dolen james dot dolen at gmail dot com

Priorityusers' Responsibility

  • Request permission to use the role
  • Create and use a proxy with the priorityuser Role when doing high priority work
    • following these instructions
    • that proxy is needed for every crab command, whenever you logout or proxy expires, do it again
    • make sure you add to crab.cfg, in the GRID section the line
      • role = priorityuser
  • Select the appropriate stage-out directory /store/group/"groupname" at one of your group's "T2_XX_YY"
    • for the correct spelling of the "groupname" follow the CMS.PhEDEx group convention
    • download GetStorageCfgFromTFC.py into a CMSSW release area
    • execute python GetStorageCfgFromTFC.py --site="T2_XX_YY" --lfn=/store/group/"groupname" (replace "T2_XX_YY" and "groupname")
    • include the script output in the USER section of your crab.cfg
  • Beware: by default all your shell sessions on the same host use the same proxy, do not try to operate at same time on normal task and priorityuser tasks (which is possible, but it is an expert operation and if you have no idea how to do it properly, it is better you do not do it). Using different roles on different machines (even simply two lxplus nodes) is OK.

Tier-2 Sites' Responsibility

In order to facilitate data transfers of physics results between Tier-2 sites, DDT has organized the commissioning of data transfer links in PhEDEx between Tier-2 sites associated with the same analysis group. Details can be found here. Priorities for link commissioning are suggested by the groups themselves. In practice, if groups have one or two Tier-2 sites with good connectivity in both directions to all of the others in a group, then data can be transfered via these "hub" Tier-2 sites between any of the other Tier-2's with at most 1 "hop" via the hub. Tier-2s should make sure that the -ignore and -accept options in their FileExport and FileDownload agents of PhEDEx are set up properly to enable transfers from other Tier-2s in their groups.

Please grant write access for VOMS CMS role=production to /store/results/ and for role=priorityuser to /store/group/. Please also check your TFC for these new LFN conventions. In addition, for data replications, PhEDEX needs to have read and write access on /store/results/.

Please configure your batch system that the share from the CMS pledges is 50% for role=production,the other half to be distributed 25% for role=priorityuser and 25% for "normal" CMS users. The appropriate way for the setup is to use a mapping to Pool Accounts in all three cases.

The StoreResults Service

Purpose of a migration from /store/group/ to /store/results/

Athough group skims or e.g. special group formats like PATs produced by a priorityuser, stored in /store/group// and registered in an analysis local-scope DBS instance can be accessed by any CMS Grid user through Crab, these samples are not official CMS data and are neither transferable to other sites using CMS.PhEDEx nor included in the groups' disk quota accounting.The StoreResults service elevated such a data set if approved by one of the performance or physics groups.

Important: Only datasets published with > CRAB_2_6_0 EDM can be migrated using the Store Results service. The files may not contain group specific data formats which are not in official CMSSW releases.

Important: At the moment growing datasets are not supported. You can not make another request for elevation with the same dataset name, if you produce more similar data you need to do something like adding _v1, _v2, _v3 etc. to the name.

Creating a StoreResults Request.

To create a StoreResults request follow these instructions

Old-Savannah conveners

Physics group Allowed to approve Savannah account Allowed ... Savannah ... Allowed ... Savannah ... Allowed to ... Savannah ... Allowed to ... Savannah ...
b-physics Hermine Woehri yes Alessandra Fanfani yes
b-tagging Wolfgang Adam yes Andrea Rizzi yes
beyond-two-generations Freya Blekman no? Salvatore Rappoccio yes? Victor Bazterra yes?
e-gamma_ecal Chris Seez yes Paolo Meridiani yes Giovanni Franzoni yes Tommaso Tabarelli de Fatisi no
ewk Markus Klute yes Michael Schmitt yes Hamel de Montch. Gautier yes
exotica Steve Worm yes John Paul Chou yes
forward Krzysztof Piotrzkowski yes Grzegorz Brona yes
heavy-ions Olga Kodolova yes Gunther Roland no
higgs Markus Klute yes Javier Fernandez yes
jets-met_hcal Victor Daniel Elvira no Philipp Schieferdecker yes Frank Chlebana yes Vladimir Gavrilov no
muon Slava Valuev yes Riccardo Bellan yes
qcd Vivian ODell yes Nikos Varelas no
susy Chris Silkworth yes Alexis Kalogeropoulos yes Jim Smith yes Dominick Olivito yes
tau-pflow Colin Bernet yes Simone Gennai yes
top Roberto Tenchini yes Roberto Chierici yes Stijn Blyweert yes Michael Maes yes Thorsten Chwalek yes
tracker-dpg Christophe Delaere yes Thomas Speer yes
tracker-pog Kevin Burkett yes Boris Mangano yes
trigger Sridhara Dasu yes Christos Leonidopoulos yes

Monitoring of the progress of the StoreResults tasks

At the moment we use two StoreResults instances at RWTH Aachen and Imperial College London, task IDs which can be divided by 3 are handled by IC, the other 2/3 by Aachen. For both instances the status of the tasks can be monitored by the priorityusers. We have imposed an access restriction based on Grid DNs given above to restrict the database queries, if you don't have access please send an Email with your DN to Manuel Giffels and Thomas Kress.

Ensure the data is located at global DBS and CMS.PhEDEx DB

First the StoreResults tool merges the /store/group/ files and stores the output, using file sizes appropriate for a performant data transfer, at the /store/results/ directory and publishes the data set in global DBS. After this step the dataset is visible by a global DBS query and can already be used for user analyses. Usually the original dataset in /store/group/ should then be removed with the help of the local Tier-2 data managers or by using Grid file deletion commands to clean the buffer space. Unfortunately the local-scope DBS dataset registration is not cleaned automatically in parallel and at the moment we can not suggest an easy way the user can clean-up the local-scope DBS entries. (NB: DBS is deprecated, in favor of DAS. An example DAS query is shown below.)
  • Important: Do not omit the StoreResults part for a Global DBS query!
  • Assuming your dataset is called
    /TTbar_1jet/top-ProcessedDSName/USER
  • DAS query example:
    dataset=/TTbar_1jet/StoreResults-ProcessedDSName/USER

Store Results also injects the dataset into the CMS.PhEDEx database, which you can query. Once this is finished after another approval step which is done centrally during the October Exercise, the dataset is available for transfers and included in the quota group accounting.

  • CMS.PhEDEx query example (Perl syntax): first "select all" or specific Tier-2s in "Nodes Shown" from "Show options" menu, then enter
    /TTbar_1jet.*/StoreResults.*

The groups should inform their users once new datasets in /store/results/ have become accessible for analysis and transfers.

Instructions for groups to use CMS.PhEDEx replication requests from Tier-2 to Tier-2

After StoreResults the dataset in /store/results/ has become an official CMS dataset. It can be transferred to other CMS Grid sites like any other dataset located in /store/data/ or /store/mc/. Groups or users can request a replication to other Tiers, which have to be approved by the corresponding Tier data managers. If you want to move the data form one Tier-2 to another, you have to first make a CMS.PhEDEx replication request followed by a deletion request for the initial site, Phedex move requests are presently not enabled among T-2s. Please note that if you replicate a /store/results/ data set from another group Y to one of your group X Tier-2 site, the LFN will be /store/results// since CMS.PhEDEx preserves the structure. This is however trasnparent to the user and for the CMS.PhEDEx group accounting.

To request a migration search for the dataset in Global DBS and follow the CMS.PhEDEx link below the dataset name.

Review status

Reviewer/Editor and Date (copy from screen) Comments
JohnStupak - 9-September-2013 Review with minor changes
Luis Contreras - 4-March-2014 Update for DBS transition

-- ThomasKress - 06-Aug-2010

Topic attachments
I Attachment History Action Size DateSorted ascending Who Comment
PNGpng StoreResultsScreenShot2.png r1 manage 138.4 K 2010-08-06 - 12:13 ManuelGiffels  
PNGpng StoreResultsScreenShot3.png r1 manage 131.8 K 2010-08-06 - 12:15 ManuelGiffels  
PNGpng StoreResultsScreenshot1.png r1 manage 86.8 K 2010-08-06 - 12:13 ManuelGiffels  
Edit | Attach | Watch | Print version | History: r60 < r59 < r58 < r57 < r56 | Backlinks | Raw View | Raw edit | More topic actions...
Topic revision: r58 - 2018-10-19 - ChristophWissing


ESSENTIALS

ADVANCED TOPICS


 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    CMSPublic All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright &© 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback