Skip to main content

JupyterLab Overview

JupyterLab in Armada Bridge's AI Studio provides interactive notebook environments where tenant users can write code, analyze data, and develop machine learning models — with direct access to GPU, CPU, or MIG (Multi-Instance GPU) resources provisioned by the Tenant Admin.

What You Can Do

  • Run Python notebooks on GPU, CPU, or MIG environments
  • Develop and test ML models with access to hardware resources
  • Verify GPU and CPU utilization with built-in scripts
  • Work in isolated, per-user environments within your tenant

Server Profiles

When adding a JupyterHub server, choose the profile that matches your workload:

ProfileResourcesUse case
Environment with GPU accessFull GPUML training, CUDA, deep learning
Environment with CPUCPU onlyData analysis, scripting, light computation
Environment with MIG GPU accessGPU partitionWorkloads that need a fraction of a GPU

Prerequisites

Before using JupyterLab:

  • A JupyterHub cluster must exist — created by the Tenant Admin using the JupyterHub with KAI Scheduler cluster template
  • You must have a tenant user account created by the Tenant Admin
  • For MIG environments, the Tenant Admin must have configured a MIG profile

In This Section