Cassandra Cloud Deployment Guide
This guide covers deploying and operating Apache Cassandra in cloud environments.
| Platform |
Guide |
Key Features |
| AWS |
Amazon Web Services |
EC2, EBS, Multi-AZ |
| GCP |
Google Cloud Platform |
GCE, Persistent Disks |
| Azure |
Microsoft Azure |
VMs, Managed Disks |
| Kubernetes |
Container orchestration |
Operators, StatefulSets |
Cloud Considerations
Instance Selection
| Workload |
AWS |
GCP |
Azure |
| Development |
m5.large |
n2-standard-2 |
Standard_D2s_v3 |
| Production |
i3.2xlarge |
n2-highmem-8 |
Standard_L8s_v2 |
| High Performance |
i3en.3xlarge |
n2-highmem-16 |
Standard_L16s_v2 |
Storage Options
Recommended storage types:
- AWS: io2 or gp3 EBS, i3 NVMe
- GCP: pd-ssd, local SSD
- Azure: Premium SSD, Ultra Disk
Requirements:
- IOPS: 3000+ for production
- Throughput: 125+ MB/s
- Latency: < 1ms
Network Configuration
# Security group rules (AWS example)
Inbound:
- Port 9042: Client connections (CQL)
- Port 7000: Inter-node communication
- Port 7001: Inter-node SSL
- Port 7199: JMX monitoring
Outbound:
- All traffic to cluster nodes
Multi-Region Deployment

Best Practices
High Availability
- Deploy across 3+ availability zones
- Use RF=3 minimum per region
- Configure proper snitch for cloud
- Enable rack awareness
Cost Optimization
- Use reserved instances for stable workloads
- Right-size instances based on actual usage
- Consider spot instances for non-critical workloads
- Monitor and optimize storage
Security
- Use VPCs/VNets for network isolation
- Enable encryption at rest and in transit
- Use IAM roles instead of credentials
- Regular security audits
Next Steps