Mongodb Atlas Setup

• Whitelist your IP address
• Or allow access from anywhere (0.0.0.0/0) for development

Cluster Tiers

M0 Sandbox (FREE - Perfect for Learning)

• 512MB storage
• Shared infrastructure
• Single region
• No backup
• Best for: Learning, development, prototyping

M2/M5 Shared Tier ($9-57/month)

• 2GB-10GB storage
• Shared infrastructure
• Global clusters available
• Daily backups
• Best for: Staging, small production apps

M10+ Dedicated Tier ($57+/month)

• 10GB+ storage (scalable)
• Dedicated servers
• All MongoDB features
• Point-in-time recovery
• Best for: Production systems

Cluster Configuration

Region Selection Strategy

Global users? → Multi-region cluster
USA only? → us-east-1 or us-west-2
EU users? → eu-west-1
Asia-Pacific? → ap-southeast-1

Network Security

// 1. IP Whitelist (RECOMMENDED)
// Add specific IPs that can access cluster
// e.g., your office, AWS security group, CI/CD runner

// 2. Public Internet (NOT RECOMMENDED)
// Allow 0.0.0.0/0 - anyone can try to connect
// Must have strong password!

// 3. VPC Peering (ENTERPRISE)
// Connect via private network
// Most secure option

Connection String Options

// Standard: Direct connection
mongodb+srv://user:[email protected]/database

// With options
mongodb+srv://user:[email protected]/database?retryWrites=true&w=majority

// Connection pooling
mongodb+srv://user:[email protected]/database?maxPoolSize=100

// TLS required
mongodb+srv://user:[email protected]/database?ssl=true

Database Users & Authentication

Create Users in Atlas

1. Security → Database Access
2. Add Database User
3. Choose authentication method:
   - Password: Username + strong password
   - Certificate: X.509 certificates
   - LDAP: Enterprise directory
4. Assign roles:
   - Atlas admin
   - Project owner
   - Editor
   - Viewer
   - Custom roles

User Roles Explained

• Project Owner: Full control of project
• Editor: Create/delete databases, manage users
• Viewer: Read-only access to project
• Database User: Application-level authentication

Backups & Disaster Recovery

Atlas Backups

Free M0: No automatic backups
M2-M5: Daily snapshots
M10+: Hourly snapshots + point-in-time recovery

Backup retention:
- 2 weeks default (M10+)
- 90 days for M10+
- Longer retention available

Restore Process

1. Atlas Console → Backup
2. Select snapshot
3. Restore to new cluster or existing
4. Wait for restore to complete
5. Verify data integrity

Backup Best Practices

1. Test Restores Regularly - Verify backups work
2. Keep Backups in Different Region - Disaster recovery
3. Export Important Data - Separate backup outside Atlas
4. Monitor Backup Size - Affects storage costs
5. Set Retention Policy - Balance cost vs. recoverability

Monitoring & Performance

Atlas Monitoring Dashboard

- Connections: Active connections, client info
- Operations: Reads, writes, inserts, updates
- CPU: CPU usage percentage
- Memory: Memory usage
- Storage: Database size growth
- Network: Incoming/outgoing bytes
- Queries: Slow queries
- Locks: Lock contention

Performance Advisor

Automatic recommendations:
- Missing indexes on frequently queried fields
- Schema improvements
- Connection pooling suggestions
- Storage optimization

Implement recommendations:
1. Review suggestion
2. Preview impact
3. Apply with one click
4. Monitor results

Alerts Configuration

Set alerts for:
- High CPU (> 75%)
- High memory (> 90%)
- Low disk space (< 10% available)
- Slow queries (> 1000ms)
- Replication lag
- Connection issues
- Long-running operations

Advanced Features

Global Clusters (Enterprise)

Multi-region replication:
- Write to any region
- Read from nearest region
- Low-latency access globally
- Automatic failover

VPC Peering

Private network connection:
- Connect from AWS VPC
- No internet exposure
- Lowest latency
- Most secure

Encryption at Rest

Enabled on M10+ by default:
- AES-256 encryption
- Managed by AWS/GCP/Azure
- Automatic key rotation
- HIPAA/SOC2 compliant

LDAP Integration

Enterprise directory authentication:
- Connect to company LDAP/Active Directory
- Centralized user management
- Automatic sync
- Role mapping

Cost Optimization

Estimate Costs

Free: $0/month (M0)
Small App: $9-57/month (M2-M5)
Production: $57-500+/month (M10-M30)

Plus data transfer costs:
- Within AWS region: Free
- Cross-region: $0.02/GB
- Internet egress: $0.03/GB

Reduce Costs

1. Use Free M0 - During development
2. Choose Right Region - Avoid data transfer costs
3. Monitor Growth - Scale appropriately
4. Compress Data - Reduces storage, network
5. Archive Old Data - Keep only active data

Troubleshooting

Connection Issues

❌ "IP not whitelisted"
✅ Add your IP to whitelist in Security → Network Access

❌ "Authentication failed"
✅ Check username/password, correct database

❌ "Cluster not available"
✅ Wait a few minutes, clusters take time to start

❌ "Connection timeout"
✅ Check firewall, increase timeout, check IP whitelist

Performance Issues

❌ Slow queries
✅ Check Performance Advisor, add indexes

❌ High CPU
✅ Review resource usage, optimize queries, upgrade tier

❌ Storage full
✅ Archive data, delete old records, upgrade tier

Best Practices

✅ Security:

1. Use strong, random passwords (20+ characters)
2. Restrict IP whitelist to specific IPs
3. Enable encryption (M10+)
4. Use separate users per application
5. Rotate passwords regularly

✅ Performance:

1. Monitor metrics regularly
2. Create indexes early
3. Test under production load
4. Use connection pooling
5. Archive old data

✅ Operations:

1. Set up alerts
2. Test backup/restore process
3. Monitor costs monthly
4. Document setup procedure
5. Plan capacity growth

Next Steps

1. Create MongoDB Atlas Account - Free tier
2. Deploy M0 Cluster - 5 minutes
3. Load Sample Data - Understand structure
4. Create Application User - Separate from admin
5. Setup Monitoring - Alerts configured
6. Plan Backup Strategy - Before going production

Ready to power your app with MongoDB in the cloud! ☁️