Getting started

Installation

1. Create a multinode account
2. Install the multinode Python library: pip install multinode
3. Log in to your account: multinode login

And you're ready to go!

Running your first distributed program on multinode

Let's run the example from the introduction. First, create a file called square.py with the following content:

# square.py

import multinode as mn

@mn.function()
def square(x):
    return x ** 2

numbers = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]

squares = square.map(numbers)
sum_of_squares = sum(squares)
print(sum_of_squares)  # 385

Now run this command:

multinode run square.py

And viola! You've just distributed your computation across ten machines in our hosted cloud environment. All in eight lines of code!

What's next?

The remaining sections of this documentation take you from our basic example to a fully-fledged application running in production.

• Core concepts - A toolkit of compute primitives that offer you the flexibility to implement diverse distributed architectures.
• Storage - Options for storing state that is shared between processes in your distributed application.
• Resource management - Customising your distributed application for your specific needs.
• Deployments - Managing the lifecycle of your application in production.