I would suggest to have a compute node view, a bit like the login node view. This would be more of an admin/staff view.
There would be two panels, one with /nodes that would have one graph with four lines one or a small number of graphs for all the nodes. The data could be: memory (%), cpu(%), localscrach used (%) and perhaps networking (Gb).
Then there would be a specific view with /nodes/<node-name>, that would have a job-like panel.
I would suggest to have a compute node view, a bit like the login node view. This would be more of an admin/staff view.
There would be two panels, one with
/nodesthat would have one graph with four lines one or a small number of graphs for all the nodes. The data could be: memory (%), cpu(%), localscrach used (%) and perhaps networking (Gb).Then there would be a specific view with
/nodes/<node-name>, that would have a job-like panel.