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.
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.
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.
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.
This project was a team effort by
| Name | GitHub Profile |
|---|---|
| Aveek Saha | aveek-saha |
| Arvind Srinivasan | arvindsrinivasan |