The real-time dashboard in Control (currently in preview release stage), provides live data in a single page to view
several aspects about the overall operation of a cluster.
Suspend a running job when a higher priority job requires computing
resources.
Drill-down
to jobs that are running on a particular HPC cluster, have been submitted to a
particular queue, or are owned by a particular user.
When suspended, a job continues to occupy system resources (memory), but is not executing
and is not charged for walltime.
Choose one of the following options to select jobs:
Select multiple jobs by enabling the selection checkbox next to the jobs
and clicking Actions.
Select a single job by clicking next to the state of the job.
Select all jobs on the current page by clicking the Select
All checkbox and clicking
Actions.
Tip: Perform a job search to filter jobs and then use the Select All
checkbox to select all the jobs on the current page that match the search
criteria. Click the reset link to deselect all
jobs.
As jobs are selected, the total number of selected jobs is displayed.
Click Suspend.
The job is suspended and computing resources (CPUs and GPUs) are freed for use
by other jobs.