Tier2 di Milano Componenti e Monitoring

1
Tier2 di MilanoComponenti e Monitoring

• Luca Vaccarossa
• Milano 14 dicembre 2007

2
User Interface (UI)

• E la macchina con i comandi per la sottomissione
  a Grid
• voms-proxy-init / grid-proxy-init
• edg-job-sumit ltfile.jdlgt
• edg-job-status ltjobidgt
• edg-job-get-output

3
User Interface (UI)

• atlfarm008.mi.infn.it
• atlfarm010.mi.infn.it
• grid008.mi.infn.it

4
Computing Element

• t2-ce-01.mi.infn.it
• Grid gateway
• PBS server (TORQUE)
• MAUI scheduler

5
Computing Element

• Il sistema batch della farm e' Torque Maui. le
  code abilitate per gli utenti locali sono
• local (max cpu time 48h, max walltime 72h)
• short (coda corta con cpu riservate, max cpu time
  40m, max walltime 2h)

6
Worker Nodes (WN)

• grid009.mi.infn.it
• grid012.mi.infn.it
• grid016.mi.infn.it
• grid017.mi.infn.it
• grid018.mi.infn.it
• grid019.mi.infn.it
• grid021.mi.infn.it
• grid022.mi.infn.it

• grid023.mi.infn.it
• grid024.mi.infn.it
• grid025.mi.infn.it
• grid026.mi.infn.it
• t2-wn-02.mi.infn.it
• t2-wn-03.mi.infn.it
• t2-wn-04.mi.infn.it
• t2-wn-05.mi.infn.it

7
Worker Nodes (WN)

• t2-wn-06.mi.infn.it
• t2-wn-07.mi.infn.it
• t2-wn-08.mi.infn.it
• t2-wn-09.mi.infn.it
• t2-wn-13.mi.infn.it
• t2-wn-14.mi.infn.it
• t2-wn-15.mi.infn.it
• t2-wn-16.mi.infn.it

• t2-wn-17.mi.infn.it
• t2-wn-18.mi.infn.it
• t2-wn-19.mi.infn.it
• t2-wn-21.mi.infn.it
• t2-wn-22.mi.infn.it
• t2-wn-23.mi.infn.it
• t2-wn-24.mi.infn.it

8
Comandi PBS

• showq
• Show job status and some job info
• showbf -v
• Check for immediately available CPUs and nodes
• checkjob -v ltjob_idgt qstat -f ltjob_idgt
• Check job status
• canceljob ltjob_idgt
• Cancel a job, sending essentially a qdel to the
  pbs_server
• showstart -h ltjob_idgt
• Show when job is scheduled to start

9
Comandi PBS

• PBSNODES a less
• Si vedono i WN che non hanno job
• Segnalare a grid-help_at_mi.infn.it

10
Priorita e FairShare

• Priorita diagnose p
• http//tier2.mi.infn.it/priorita.txt
• FS diagnose f
• http//tier2.mi.infn.it/fairshare.txt

11
Chi sono io ?

• "/CIT/OINFN/OUPersonal Certificate/LMilano/CN
  Silvia Resconi/EmailSilvia.Resconi_at_mi.infn.it"
  resconi
• "/CIT/OINFN/OUPersonal Certificate/LMilano/CN
  Tommaso Lari" lari

12
Chi sono io ?

• "/CIT/OINFN/OUPersonal Certificate/LMilano/CN
  Attilio Andreazza" andreazz
• "/CIT/OINFN/OUPersonal Certificate/LMilano/CN
  Clara Troncon" troncon
• "/CIT/OINFN/OUPersonal Certificate/LMilano/CN
  Leonardo Carminati" lcarmina

13
Chi sono io ?

• "/CIT/OINFN/OUPersonal Certificate/LMilano/CN
  Donatella Cavalli" cavalli
• "/CIT/OINFN/OUPersonal Certificate/LMilano/CN
  Caterina Pizio" pizio

14
Chi sono io ?

• "/CIT/OINFN/OUPersonal Certificate/LMilano/CN
  Umberto De Sanctis" atlas012
• "/CIT/OINFN/OUPersonal Certificate/LMilano/CN
  Simone Montesano" atlas020

15
Chi sono io ?

• "/CIT/OINFN/OUPersonal Certificate/LMilano/CN
  Chiara Tamarindi" atlas033
• "/CIT/OINFN/OUPersonal Certificate/LGenova/CN
  Fabrizio Parodi" parodi
• "/CIT/OINFN/OUPersonal Certificate/LGenova/CN
  Bianca Osculati" osculati

16
GridView

• http//gridview.cern.ch/GRIDVIEW/
• Monitoring and Visualization Tool for LCG
• Data Transfer
• Job Status  
• Service Availability

17
(No Transcript)
18
SAM Tests

• https//lcg-sam.cern.ch8443/sam/sam.py
• Certificato nel browser
• Test automatici
• SAM on demand?
• https//cic.gridops.org/index.php?sectionrcpage
  samadmin

19
(No Transcript)
20
(No Transcript)
21
Ganglia

• http//ganglia.sourceforge.net/
• Ganglia is a scalable distributed monitoring
  system for high-performance computing systems
  such as clusters and Grids.

22
Ganglia

• It relies on a multicast-based listen/announce
  protocol to monitor state within clusters and
  uses a tree of point-to-point connections amongst
  representative cluster nodes to federate clusters
  and aggregate their state.
• It leverages widely used technologies such as XML
  for data representation, XDR for compact,
  portable data transport, and RRDtool for data
  storage and visualization.