Run multiple instances of a python application

Cluster information:

Kubernetes version: Latest
Cloud being used: (put bare-metal if not on a public cloud) bare metal
Host OS: Ubuntu 16.04

I am a newbie to kubernetes so if responses are a little detailed it would be highly appreciated.

My objective is to run multiple instances of a python application. Each instance will be passed with static number of arguments (5 arguments). Each instance will basically run ffmpeg decoding in GPU and generate kakfa streams. Trigger to create a new instance will come on socket.io.

I would like to manage the process state, maintain separate logs (log file name would be an argument). Would also like to seek help on how to start and stop the process. Would require help on how to achieve this.

This is my first post so please be patient if any guidelines have not been followed.