Welcome to the Mystic Challenge

Your goal is to create a service that can receive user-provided functions (see example function below) remotely and distribute their execution amongst multiple workers. Users can interact with the service via a Python client you should also create. Example function:

import time

def func(x):
    time.sleep(x)
    return x*x

The client-side API to run the above function on the service should look like:

import time

from distribute_challenge import compute_this

@compute_this()
def func(x):
    time.sleep(x)
    return x*x
    
out = func(2).run()
assert out == 4

Deliverable

• Write the Python module distribute_challenge which provides the client-side API above.
• Build the API service that allows remote function submission. Submitted functions should be distributed to workers for execution and the results returned to the client.
  • The number of workers should be configurable when starting the service.
  • The user function is sent to a single worker. Workers run a single user function at a time.
• Load test the API and demonstrate with a monitoring system of your choice throughput and allocation to each worker. Measure how your system scales with the number of workers and functions.
• You can use open source libraries to help you solve the challenge, however the more of your own skills you show the better! Note: Ray cannot be used.

Architecture diagram of deliverable