Container Orchestration System for a Scalable API

An Orchestration System for a containerised API broken into microservices.

Features a generic orchestrator that can be used with any application based on custom scaling and health checking rules.

Orchestration features:

Fault tolerance :

Every 2 seconds, the orchestrator engine polls the health check API of each running Acts container. On detecting an unhealthy container, it restarts that container on the same port.

Auto Scaling:

The orchestrator keeps a track of the number of incoming HTTP requests in the past two minutes. At every 2 minute interval, depending on how many requests were received, the orchestrator increases or decreases the number of containers.

Load Balancing:

The round robin scheduling algorithm, that we have used to distribute requests, is a scheduling algorithm where each process is assigned a fixed time slot in a cyclic way.

Team

This project was a team effort by

Name	GitHub Profile
Aveek Saha	aveek-saha
Arvind Srinivasan	arvindsrinivasan