Skip to content

Quickstart

This guide shows you how to use the Jumper Wrapper Kernel in a few simple steps.

Step 1: Select the Kernel

  1. Start Jupyter Notebook or JupyterLab
  2. Create a new notebook
  3. Select Jumper Wrapper Kernel as your kernel

Step 2: List Available Kernels

See which kernels can be wrapped:

%list_kernels

Output: 

Available Jupyter Kernels:
--------------------------------------------------
  python3: Python 3 (ipykernel) (python)
  ir: R (r)
  julia-1.9: Julia 1.9 (julia)
--------------------------------------------------

Step 3: Wrap a Kernel

Wrap your desired kernel:

%wrap_kernel python3

Output: 

Successfully wrapped kernel: python3
Hint: Refresh the page (without restarting the kernel) to enable syntax highlighting for the wrapped language.

Step 4: Use Performance Monitoring

Now you can use jumper-extension commands while running code on the wrapped kernel:

# Start monitoring (handled locally)
%perfmonitor_start

# Run code on the wrapped kernel
import numpy as np
x = np.random.rand(1000, 1000)
y = np.dot(x, x.T)

# View performance report (handled locally)
%perfmonitor_perfreport

How It Works

The Jumper Wrapper Kernel acts as a proxy:

  1. Magic commands from jumper-extension are intercepted and executed locally
  2. All other code is forwarded to the wrapped kernel
  3. Output is streamed back to the notebook

This allows you to monitor performance of any Jupyter kernel, regardless of its language.