Skip to content

Workload Management

A workload is a collection of containers (e.g., a Kubernetes Deployment, DaemonSet, or StatefulSet) that run a particular application or service inside a workspace. Most commonly a workspace and workload has a one-to-one relationship which simplifies management. When you start a Workspace, you effectively launch its associated workload, and stopping or deleting the Workspace likewise terminates the underlying workload. For example, interactive workloads such as Jupyter Notebook or VSCode environments.

Viewing and Controlling Workloads

Open the “Workloads” Dashboard
- In the left navigation bar, select the Workloads Icon Workloads icon.
- You will land on the Workspaces dashboard, which shows all current Workspaces.
- If no Workspaces exist yet, the page indicates “No workspace to display.”

Workloads Dashboard

The screen presents a clear table of your workloads, along with tools to help you find and understand them.

Workload Management Controls

  • Search Workloads Field: Easily find a specific workload by typing its name into this search bar.

  • All Projects Dropdown: Filter the list of workloads to show only those belonging to a specific project you select.

  • Workload Status Summary: A quick overview of workloads categorized by their current state:

  • All: Total count of all your workloads.

  • Pending: Workloads that are waiting to be scheduled or started.
  • Running: Workloads currently active and executing (e.g., 1 in the example).
  • Succeeded: Workloads that have completed successfully (e.g., 0).
  • Failed: Workloads that terminated with an error (e.g., 0).
  • Evicted: Workloads that were removed from a node (e.g., due to resource contention).
  • Preempted: Workloads that were stopped to free up resources for higher-priority tasks.

Workloads List Table

Each row in the table represents an individual workload and provides the following key details:

  • Name: The name of your workload (e.g., workspace-ws-purple-5d7f8). Clicking on the name will typically take you to a detailed view of that specific workload.

  • Status: The current operational state of your workload (e.g., Running).

  • Priority: The priority level assigned to the workload (e.g., High).

  • Project: The project to which this workload belongs (e.g., proj-blue).

  • Node Groups: The number of node groups on which this workload is running or scheduled to run (e.g., 1).

  • Requested Resources: The amount of memory, CPU, and GPU resources allocated to this workload (e.g., MEM: 1GiB | CPU: 1 Cores | GPU: 1).

  • Created At: The date and time when the workload was originally created or submitted (e.g., 2025-06-13).

  • Actions: Icons that allow you to perform actions on the workload:

  • Delete Icon (Trash Can): Permanently removes the workload entry. Use with caution, as this typically deletes associated data or stops the running job.

Pagination Controls

  • At the bottom right of the table, navigation controls (e.g., 1 / 10 / page) allow you to move between pages if you have many workloads, and adjust how many workloads are displayed per page.

Workloads Dashboard Benefits

  • Monitor Workload Status: Quickly see the current state of all your submitted workloads.
  • Track Progress: Identify which workloads are Running, Pending, or have completed.
  • Review Resource Usage: See the resources requested by each of your workloads.
  • Manage Workloads: Delete workloads that are no longer needed.
  • Access Workload Details: Click on a workload's name for a more in-depth view and additional management options.

Best Practices

  • Use Projects: Assign your workloads (and hence Workspaces) to a relevant Project for logical grouping and easier management.
  • Monitor Status: Keep an eye on workload statuses in case of errors (e.g., CrashLoopBackOff) that might require adjustments to your container image or resource limits.
  • Resource Efficiency: Stop or delete workloads that are no longer active to free cluster resources (GPUs, CPUs, memory).
  • Persistent Data: Attach volumes to any workload-based session (Workspace) that needs to retain data beyond the lifecycle of the container.