This is part of #66 and also closely related to #566, which needs to have the same functionality. When available on a system, the Slurm data provide valuable insights into the behavior of a job: requested / allocated resources, used resources, queue time, completion code, and so on - everything we can get from sonalyze sacct when it is done. These data need to be made available in the Job Query dashboard, and in particular, if eg resource utilization is bad there should probably be something highlighting that so that it's quite visible.
I think this is a difficult thing to do because data overload is a real thing, but we should grab a whiteboard and sketch out some possibilities.
There is also the further possibility that a job that is a policy violator should be flagged as such in the Job Query view, but this will take a little bit more doing on the back end.
This is part of #66 and also closely related to #566, which needs to have the same functionality. When available on a system, the Slurm data provide valuable insights into the behavior of a job: requested / allocated resources, used resources, queue time, completion code, and so on - everything we can get from
sonalyze sacctwhen it is done. These data need to be made available in the Job Query dashboard, and in particular, if eg resource utilization is bad there should probably be something highlighting that so that it's quite visible.I think this is a difficult thing to do because data overload is a real thing, but we should grab a whiteboard and sketch out some possibilities.
There is also the further possibility that a job that is a policy violator should be flagged as such in the Job Query view, but this will take a little bit more doing on the back end.