ATLAS Batch monitoring

C5 report

• C5 reports: https://monit-grafana.cern.ch/d/000000859/c5-summary?orgId=5
  • Howto:
    • Select C5 Summary dashboard

Experiment view

Let's call this the main Experiment view...

• Experiment view: https://monit-grafana.cern.ch/d/000000865/experiment-batch-details?orgId=5&var-cluster=cernprod&var-experiment=ATLAS

  • Howto:
    • Select Experiment Batch Detail dashboard
    • Select cluster: cernprod
    • Select Experiment: ATLAS

• Quota plot: https://monit-grafana.cern.ch/d/000000865/experiment-batch-details?orgId=5&var-cluster=cernprod&var-experiment=ATLAS&panelId=18&fullscreen

  • Howto:
    • Start from the "main Experiment view", i.e. select Experiment Batch Detail dashboard, select Experiment: ATLAS
    • Click on the "New Utilization (Cpus) of quota share [%]" plot next to plot title, select View.

• Queued jobs: https://monit-grafana.cern.ch/d/000000865/experiment-batch-details?orgId=5&var-cluster=cernprod&var-experiment=ATLAS&panelId=22&fullscreen

  • Howto:
    • Start from the "main Experiment view", i.e. select Experiment Batch Detail dashboard, select Experiment: ATLAS
    • Click on the "Queued jobs on central manager, by Experiment, AccountingGroup and RequestedCpus" plot next to plot title, select View.

• Running jobs: https://monit-grafana.cern.ch/d/000000865/experiment-batch-details?orgId=5&var-cluster=cernprod&var-experiment=ATLAS&panelId=23&fullscreen

  • Howto:
    • Start from the "main Experiment view", i.e. select Experiment Batch Detail dashboard, select Experiment: ATLAS
    • Click on the "Running jobs on central manager, by Experiment, AccountingGroup and RequestedCpus" plot next to plot title, select View.

Condor T0 cluster

Let's call this the main Experiment T0 view...

• Condor T0: https://monit-grafana.cern.ch/d/000000865/experiment-batch-details?orgId=5&panelId=23&var-cluster=cernt0&var-experiment=ATLAS

  • Howto:
    • Select Experiment Batch Detail dashboard
    • Select cluster: cernt0
    • Select Experiment: ATLAS

• Queued jobs on T0: https://monit-grafana.cern.ch/d/000000865/experiment-batch-details?orgId=5&var-cluster=cernt0&var-experiment=ATLAS&panelId=22&fullscreen

  • Howto:
    • Start from the "main Experiment view", i.e. select Experiment Batch Detail dashboard, select Experiment: ATLAS
    • Click on the "Queued jobs on central manager, by Experiment, AccountingGroup and RequestedCpus" plot next to plot title, select View.

• Running jobs on T0: https://monit-grafana.cern.ch/d/000000865/experiment-batch-details?orgId=5&var-cluster=cernt0&var-experiment=ATLAS&panelId=23&fullscreen

  • Howto:
    • Start from the "main Experiment view", i.e. select Experiment Batch Detail dashboard, select Experiment: ATLAS
    • Click on the "Running jobs on central manager, by Experiment, AccountingGroup and RequestedCpus" plot next to plot title, select View.

BEER

monitoring in development.

This has ATLAS and LHCb together. For ATLAS only we need to enrich probes.

• ATLAS + LHCb BEER: https://monit-grafana.cern.ch/d/000000867/occupancy?orgId=5&var-cluster=cernprod&var-datacentre=meyrin&var-datacentre=wigner&var-hostgroup=eos
  • Howto:
    • Select Occupancy dashboard, select hostgroup: eos

External cloud

perhaps this is a experimental dashboard, may or may not change in the future...

It contains plots by experiment for TSY/RHEA.

• External cloud: https://monit-grafana.cern.ch/d/sGS0_cIik/ben-xbatch?orgId=5
  • Howto:
    • Select Ben xbatch dashboard

Users

Batch documentation for users: https://batchdocs.web.cern.ch/

User activity monitoring is available by username, only for local submission (i.e. not grid jobs), e.g.:

• atltzp1 last 60 days: https://monit-grafana.cern.ch/d/000000869/user-batch-jobs?refresh=5m&orgId=5&var-cluster=cernt0&var-user=atltzp1&from=now-60d&to=now

---

Last update: March 20, 2024