CryptoDash – Market Server (Backend)

CryptoDash Server is a Node.js backend that fetches cryptocurrency market data from Binance, stores it in a database, computes technical indicators, and exposes a REST API for visualization clients.

This server is designed to be indicator-first: all heavy calculations (ATR, Supertrend, etc.) are done server-side.

---

Features

• Automated candle fetching from Binance
• OHLCV data storage (per symbol & timeframe)
• ATR (Wilder) calculation
• Supertrend indicator calculation
• Incremental updates (no refetching full history)
• Cron-based scheduler aligned with candle closes
• REST API for frontend dashboards

---

Tech Stack

• Node.js
• Prisma ORM
• PostgreSQL (or SQLite for local testing)
• Binance REST API
• node-cron

---

Project Structure

src/
├─ api/
│  └─ candles.js
├─ services/
│  ├─ binance.js
│  ├─ candleStore.js
│  └─ indicators.js
├─ cron/
│  └─ scheduler.js
├─ config/
│  └─ market.js
├─ prisma/
│  └─ schema.prisma
└─ index.js

---

⚙️ Setup & Run (Local)

1️⃣ Install dependencies

npm install

---

2️⃣ Environment variables

Create .env file:

DATABASE_URL=postgresql://user:password@localhost:5432/cryptodash
BINANCE_API_KEY=
BINANCE_API_SECRET=
PORT=4000

---

3️⃣ Database setup

npx prisma migrate dev

---

4️⃣ Start server

npm run dev

Server runs at:

http://localhost:4000

---

Scheduler

The server runs scheduled jobs aligned with Binance candle closes:

• 5m → every 30 seconds (forming candle updates)
• 15m → every 5 minutes
• 1h → hourly
• 4h → every 4 hours
• 1d → daily

Each job:

1. Fetches latest candles
2. Stores / updates last candle
3. Recalculates ATR
4. Recalculates Supertrend

---

API Endpoints

Get candles

GET /api/candles?symbol=BTCUSDT&tf=1h&limit=200

Response

{
  "candles": [
    {
      "openTime": "2024-01-01T00:00:00.000Z",
      "open": 42000,
      "high": 42500,
      "low": 41800,
      "close": 42350,
      "volume": 1234,
      "takerBuyVolume": 650,
      "atr": 320,
      "supertrend": 41750,
      "supertrendDir": 1
    }
  ]
}

---

Indicator Logic

ATR

• True Range calculation
• Wilder smoothing
• Stored per candle

Supertrend

• Uses HL2 ± ATR × multiplier
• Sticky bands
• Direction state persisted
• Stored directly in candle records

---

Important Notes

• Indicators depend on historical continuity
• Deleting candles may break indicator state
• Cron jobs should not overlap
• SQLite is fine for local testing, PostgreSQL recommended for real usage

---

Disclaimer

This project is for educational and research purposes only.
It is not financial advice.

---

Author

VanHes1ng
Crypto systems & market analytics