Fibonacci Sequence

Short description

This project is a microservices-based Fibonacci calculator deployed in a Kubernetes cluster. The system accepts an index from a React frontend, forwards it to an Express API, and calculates the Fibonacci value for that index in a distributed, fault-tolerant way.

---

Table of contents

1. Architecture
2. Technologies
3. Repository layout
4. Local development
5. Docker & Kubernetes
6. Environment variables

---

Architecture

High-level components:

• React (client) — single-page app where user submits an index.

• API (Express) — receives index submissions and exposes API endpoints.

  • Writes received indices to Postgres (permanent history).
  • Pushes the index into Redis (as a job or key placeholder) for processing.

• Worker — a separate process that watches Redis for new indices, computes the Fibonacci value, and writes the result back into Redis (keyed by index).

• Redis — used as ephemeral store/cache and a simple task queue between the API and worker.

• Postgres — persistent store of all received indices (audit/history).

• Kubernetes — manifests for deploying client, server, worker, Redis, and Postgres.

---

Technologies

• Frontend: React
• Backend: Node.js, Express
• Worker: Node.js
• Cache/Queue: Redis
• Database: PostgreSQL
• Containerization: Docker
• Orchestration: Kubernetes (manifests located in /k8s)
• CI/CD: (suggested) GitHub Actions / GitLab CI

---

Repository layout

microservices/
├─ client/                  # React app
├─ server/                  # Express API
│  ├─ worker/               # Worker that watches Redis and calculates fib
│  └─ src/
├─ k8s/                     # Kubernetes manifests (Deployments, Services, ConfigMaps, Secrets)
├─ docker-compose.yml       # Optional: compose for local dev
├─ README.md                # <- this file
└─ docs/
   ├─ architecture-diagram.png
   └─ screenshot-frontend.png

---

Local development

Prerequisites

• Docker & Docker Compose
• Node.js >= 18 (for local dev of client/server)
• kubectl (optional, for k8s testing)

Quick start (using docker-compose)

2. From project root run (starts Postgres, Redis, server, worker, client):

docker compose up --build

3. Open the client in your browser (by default http://localhost:3000).

Local (without docker)

Run Postgres and Redis locally (or use Docker for them), then:

# server
cd server
npm install
npm run dev

# worker (in a separate shell)
cd server/worker
npm install
npm run dev

# client
cd client
npm install
npm start

---

Kubernetes

Build images

# from repo root (example names)
docker build -t your-dockerhub-username/fibonacci-client:latest ./client
docker build -t your-dockerhub-username/fibonacci-server:latest ./server
docker build -t your-dockerhub-username/fibonacci-worker:latest ./server/worker

Push to your container registry and update the images in k8s/ manifests.

Deploy to Kubernetes (example)

kubectl apply -f k8s/

The k8s/ folder should contain:

• client-deployment.yaml
• client-cluster-ip-service.yaml
• server-deployment.yaml
• server-cluster-ip-service.yaml
• worker-deployment.yaml
• redis-deployment-.yaml
• redis-cluster-ip-service.yml
• postgres-deployment.yaml
• postgres-cluster-ip-service.yml
• database-persistent-volume-claim.yml
• ingress-service.yml

Notes:

• Create your own secret object as imperative.

kubectl create secret generic <secret name> --form-literal <key=pass>

---

Environment variables

Example (server .env)

PORT=5000
PG_HOST=postgres
PG_USER=postgres
PG_PASSWORD=postgrespassword
PG_DATABASE=fibdb
REDIS_HOST=redis
REDIS_PORT=6379
SECRET_KEY=some-secret

Example (client .env)

REACT_APP_API_URL=http://localhost:5000

---