Neon Serverless Postgres

When to Use This Skill

• Setting up Neon database connections
• Using database branching for development
• Configuring connection pooling
• Optimizing for serverless environments
• Integrating Neon with frameworks

Quick Start

Create Project

# Using Neon CLI
npm install -g neonctl
neonctl auth

# Create project
neonctl projects create --name my-project

# Get connection string
neonctl connection-string

Connection String Format

postgresql://[user]:[password]@[endpoint].neon.tech/[database]?sslmode=require

Connection Patterns

Serverless Function (Recommended)

// Use Neon serverless driver for edge/serverless
import { neon } from '@neondatabase/serverless';

const sql = neon(process.env.DATABASE_URL!);

export async function getUsers() {
  const users = await sql`SELECT * FROM users LIMIT 10`;
  return users;
}

Connection Pooling with Drizzle

import { drizzle } from 'drizzle-orm/neon-http';
import { neon } from '@neondatabase/serverless';

const sql = neon(process.env.DATABASE_URL!);
export const db = drizzle(sql);

// Usage
const users = await db.select().from(usersTable).limit(10);

Node.js with pg (Long-running)

import { Pool } from 'pg';

const pool = new Pool({
  connectionString: process.env.DATABASE_URL,
  ssl: { rejectUnauthorized: true },
  max: 10,
  idleTimeoutMillis: 30000,
  connectionTimeoutMillis: 10000,
});

// Use pooled connections
const client = await pool.connect();
try {
  const result = await client.query('SELECT * FROM users');
  return result.rows;
} finally {
  client.release();
}

Prisma Integration

// prisma/schema.prisma
datasource db {
  provider  = "postgresql"
  url       = env("DATABASE_URL")
  directUrl = env("DIRECT_URL") // For migrations
}

// For edge/serverless
import { PrismaClient } from '@prisma/client'
import { PrismaNeon } from '@prisma/adapter-neon'
import { neon } from '@neondatabase/serverless'

const sql = neon(process.env.DATABASE_URL!)
const adapter = new PrismaNeon(sql)
const prisma = new PrismaClient({ adapter })

Database Branching

Create Development Branch

# Create branch from main
neonctl branches create --name dev --parent main

# Get branch connection string
neonctl connection-string --branch dev

# List branches
neonctl branches list

# Delete branch
neonctl branches delete dev

Branch Workflow

## Development Workflow

1. **Main Branch**: Production data
2. **Dev Branch**: Daily development
3. **Feature Branches**: Per-feature isolation

main (production) ├── dev (shared development) │ ├── feature/auth │ └── feature/payments └── staging (pre-production)

### CI/CD Branch Strategy

```yaml
# .github/workflows/preview.yml
name: Preview Environment
on:
  pull_request:
    types: [opened, synchronize]

jobs:
  create-preview:
    runs-on: ubuntu-latest
    steps:
      - name: Create Neon Branch
        uses: neondatabase/create-branch-action@v4
        with:
          project_id: ${{ secrets.NEON_PROJECT_ID }}
          branch_name: preview-${{ github.event.pull_request.number }}
          api_key: ${{ secrets.NEON_API_KEY }}

      - name: Run Migrations
        run: npx prisma migrate deploy
        env:
          DATABASE_URL: ${{ steps.create-preview.outputs.db_url }}

Autoscaling Configuration

Compute Settings

// Neon autoscales based on load
// Configure min/max compute units in dashboard

// For predictable workloads, set minimum compute
// Project Settings > Compute > Minimum compute size

// Scale to zero saves costs during inactivity
// Re-activation takes ~500ms (cold start)

Handle Cold Starts

// Warm connection on app start
async function warmDatabase() {
  const sql = neon(process.env.DATABASE_URL!);
  await sql`SELECT 1`;
}

// Call during initialization
warmDatabase().catch(console.error);

Best Practices

Query Optimization

-- Use EXPLAIN ANALYZE
EXPLAIN ANALYZE SELECT * FROM users WHERE email = 'test@example.com';

-- Add appropriate indexes
CREATE INDEX idx_users_email ON users(email);

-- Use partial indexes for common filters
CREATE INDEX idx_active_users ON users(created_at)
WHERE status = 'active';

Connection Management

// DON'T: Create new connection per request
async function handleRequest() {
  const pool = new Pool({ connectionString: url }); // BAD
  const result = await pool.query('SELECT 1');
  await pool.end();
}

// DO: Reuse connection pool
const pool = new Pool({ connectionString: url }); // Create once

async function handleRequest() {
  const result = await pool.query('SELECT 1');
  return result;
}

Serverless Considerations

// For serverless: Use HTTP-based driver
import { neon } from '@neondatabase/serverless';
const sql = neon(process.env.DATABASE_URL!);

// For long-running: Use WebSocket for better performance
import { neonConfig, Pool } from '@neondatabase/serverless';
import ws from 'ws';

neonConfig.webSocketConstructor = ws;
const pool = new Pool({ connectionString: url });

Environment Setup

Next.js

# .env.local
DATABASE_URL="postgresql://user:pass@ep-xxx.region.aws.neon.tech/dbname?sslmode=require"

# For Prisma migrations (bypasses pooler)
DIRECT_URL="postgresql://user:pass@ep-xxx.region.aws.neon.tech/dbname?sslmode=require"

Vercel Integration

# Connect Neon to Vercel
# 1. Go to Neon Dashboard > Integrations
# 2. Select Vercel
# 3. Authorize and select project
# Environment variables are set automatically

Monitoring

Built-in Metrics

## Available in Neon Console

- Connection count
- Query performance
- Storage usage
- Compute hours
- Data transfer

Query Performance

-- Enable query statistics
CREATE EXTENSION IF NOT EXISTS pg_stat_statements;

-- Find slow queries
SELECT
  query,
  calls,
  mean_exec_time,
  total_exec_time
FROM pg_stat_statements
ORDER BY total_exec_time DESC
LIMIT 10;

Schema Management

Migrations with Prisma

# Development
npx prisma migrate dev --name add_users

# Production
npx prisma migrate deploy

Migrations with Drizzle

# Generate migration
npx drizzle-kit generate:pg

# Apply migration
npx drizzle-kit push:pg

Checklist

• Use serverless driver for edge functions
• Configure connection pooling appropriately
• Set up database branching for dev/staging
• Add indexes for common queries
• Monitor query performance
• Use SSL connections (sslmode=require)
• Configure autoscaling limits
• Handle cold starts gracefully