What's a Plexus Workload
Plexus manages compute resources through workloads, which correspond to Slurm jobs on the underlying cluster.
What is a Slurm job?
A Slurm job is a unit of work submitted to a Slurm-managed cluster. It typically consists of a script or command plus resource requirements (CPUs, GPUs, memory, runtime). Slurm schedules the job on compute nodes that meet those criteria. Jobs can run in batch or interactive mode; you can monitor them via Slurm commands or the platform interface.
Submitting jobs to a Slurm cluster directly requires SSH access to the cluster.
What is a Plexus workload?
A Plexus workload is a unit of work you submit to a Slurm or Kubernetes cluster. For each application, Plexus provides a default resource configuration that you can customize. Workloads can run in batch (non-interactive) or interactive mode.
Plexus allocates the resources, runs the application, and monitors progress. You can interact with running workloads through the platform tools.
In short, a workload is the execution of an application in an isolated environment on a cluster.
With Plexus, you do not need direct access to the underlying clusters. One Plexus account gives you access to all supported compute resources.