Monitoring with TIG Stack in the GWDG

Event content

GWDG is offering different methods to do a job analysis in regards to compute perfomance/IO and more. Besides tools that have to be started exclusively like Vampyr, the infrastructure itself offers tools which collect data continuously on the compute nodes and can correlate it to jobs. This data is offered to the users via a frontend utilizing the software Grafana, a web-based visualization tool.

This course offers a general overview of monitoring in HPC in order to allow the participants to understand how the systems interact and how data is acquired. Furthermore it gives an introduction to the usage of Grafana to analyse the collected data of the users‘ own jobs.

Learning goal

• Get an overview of monitoring in HPC at the GWDG (What is it? Why?)
• Understanding what ProfiT-HPC and Grafana is and what is used for
• Basic knowledge on Grafana usage (login, check jobs on the dashboard)

Information about the event

Max. participants

40

Requirements

None

Speakers

Marcus Merz