Skip to content

What's Granted in Plexus

Plexus manages compute resources using queues, which are based on Slurm partitions.

What is a queue in Plexus?

A Plexus queue is a logical grouping of compute nodes, derived from the Slurm partition concept. Each queue (partition) has:

  • A specific set of compute nodes with defined hardware characteristics (e.g., GPU/CPU type, memory)
  • Resource limits and scheduling policies (e.g., workload limits, time limits)
  • Intended use cases (e.g., AI/ML, HPC, data analytics)
  • Access controls for different user groups or teams

When you submit workloads to a queue, Slurm schedules them on the nodes assigned to that queue and applies the configured constraints. This helps manage cluster resources for different workloads and teams.

How queues are granted

You are assigned to a team that has access to a set of queues. When you create a workload, you choose from your available queues; Plexus then prepares the environment and resources for the selected queue.

Tips

  • Choose the queue that best matches your workload’s resource needs.
  • Review queue limits before submitting large or long-running workloads.