Adapted from gap analysis doc: https://docs.google.com/document/d/1Oz0rvy9BtTkB8FrQtamMlIMOiDrM2Ov5pUTZDzNtKXQ/edit?tab=t.0

Time-bound job scheduling in Slurm allows for better binpacking and anticipation of future system use

• Use Case: a job that starts needs to stage a lot of data, which is costly; if it gets preempted by a higher priority job, all is lost
• Cannot natively schedule based on job duration limits or predict resource availability windows, preventing efficient backfill scheduling and resource planning
• https://github.com/kubernetes-sigs/kueue/issues/6587