Cargando…
Experiment Dashboard: monitoring system for the LHC experiments
LHC experiments are depending on the distributed EGEE infrastructure for their core activities. The Experiment Dashboard is a monitoring framework aiming to provide for the LHC experiments the overview of their activities on the EGEE infrastructure with a special emphasis in support for the user com...
Autores principales: | , , , , , , , , |
---|---|
Lenguaje: | eng |
Publicado: |
2007
|
Materias: | |
Acceso en línea: | http://cds.cern.ch/record/1120913 |
Sumario: | LHC experiments are depending on the distributed EGEE infrastructure for their core activities. The Experiment Dashboard is a monitoring framework aiming to provide for the LHC experiments the overview of their activities on the EGEE infrastructure with a special emphasis in support for the user community. Existing monitoring tools are usually focusing on a specific usage like specific Grid middleware/infrastructures, specific submission tool, etc. The Experiment Dashboard has been built to aggregate existing monitoring infrastructures (from experiment specific software, infrastructure itself, monitoring tools) and provide unified views and information correlation. Experiment Dashboard is covering different areas of the LHC activities - job processing, data transfer, and data publishing. It is deployed for four LHC experiments (CMS, ATLAS, LHCb, ALICE). Some of the core functionality of the Experiment Dashboard like job monitoring can be used for other virtual organizations. Experiment Dashboard is currently in production and is used by LHC users with different roles for their everyday work. The whole EGEE monitoring infrastructure can be considerably improved. Very often the error messages indicating various failures are not clear and do not point to the real problem. The variety of the local fabric monitoring systems used by sites complicates the task of creation of the common framework for aggregation of the monitoring data in the central repository. Transparent navigation of the monitoring data provided by different monitoring systems is often not possible. |
---|