Distributed/Microservices Prime Numbers Counter

---

This project was developed for explore following areas and technologies:

• Technologies:

  • Golang

    • Golang channels

    • Client-go

  • gRPC

    • Protobuf

  • Typescript

  • Docker

  • Kubernetes

• Areas

  • Distributed computing

  • Microservices

Prerequisites:

To run this project you will be needing a Kubernetes ready cluster, you can use AKS or GKE also if you want to test it locally you can run it one minikube (a local kubernetes cluster for testing purposes). Also you should be familiar with Docker commands and kubectl Just for a sample, I'm stating two basic commands.

To create Docker image docker built -t <registry server address>/<image_name>:<version> <directory path to Dockerfile>

To create a something from a file in kubectl kubectl create -f <filename>.yaml

Don't forget to apply the deployments from preDeployments folder

Working:

This repository contains 3 total projects Client (Typescript), Provisioner (Golang) and Worker (Typescript). Basic idea of this project is, Client is a web-based application from where a client send a request to count prime numbers between a range, the request is sent to provisioner via gRPC. Provisioner divides the range in chunks if needed and launches a batch of workers and initiate some channels to listen for workers of this specific request. Workers calculate primes in their chunked range and responds back to provisioner. When all the workers for a specific request has responded, the provisioner sends the total result to Client.

A diagram shows an overall view of working of this project:

Details about each service:

• Each service communicates via gRPC

• Each service is Dockerized to run as container in Kubernetes

Some technical details are below

1. Client

   • Written in TypeScript

   • Using Express Server and EJS Views

   • Using gRPC to communicate with Provisioner

   • DockerFile included

2. Provisioner

   • Written in Golang

   • Using gRPC to communicate with Client and Worker

   • Using Golang Channels to handle the data received from workers, totaling and sending the result to client

   • DockerFile included

3. Worker

   • Written in TypeScript

   • Using gRPC to communicate with Provisioner

   • DockerFile included

---

Note: This project is solemnly for learning purposes that how you can receive a task from a client and divide that in batch processes and easily communicate via gRPC and also to demonstrate the usage of Golang channels and gRPC.