| name | cloud |
| description | Cloud platforms skill - AWS, GCP, Azure core services and architecture |
| version | 1.0.0 |
| sasmp_version | 1.3.0 |
| input_schema | [object Object] |
| output_schema | [object Object] |
| retry_config | [object Object] |
| timeout_ms | 60000 |
Cloud Platforms Skill
PURPOSE
Cloud architecture and services across major providers.
CORE SERVICES
AWS
Compute: EC2, Lambda, ECS, EKS
Storage: S3, EBS, EFS
Database: RDS, DynamoDB, ElastiCache
Networking: VPC, ALB, CloudFront, Route53
GCP
Compute: Compute Engine, Cloud Run, GKE
Storage: Cloud Storage, Persistent Disk
Database: Cloud SQL, Firestore, Memorystore
Networking: VPC, Cloud Load Balancing, Cloud CDN
Azure
Compute: VMs, Container Apps, AKS
Storage: Blob Storage, Disk
Database: SQL Database, Cosmos DB
Networking: VNet, Application Gateway
ARCHITECTURE PATTERNS
3-Tier Web App
┌─────────────────────────────────────┐
│ CloudFront/CDN │
├─────────────────────────────────────┤
│ Application Load Balancer │
├─────────────────────────────────────┤
│ Auto Scaling Group (EC2/ECS) │
├─────────────────────────────────────┤
│ RDS Multi-AZ + ElastiCache Redis │
└─────────────────────────────────────┘
COST OPTIMIZATION
Compute:
├── Reserved/Committed instances
├── Spot/Preemptible for batch
├── Right-sizing
└── Auto-scaling
Storage:
├── Lifecycle policies
├── Tiered storage
└── Delete unused
Network:
├── NAT gateway sharing
├── VPC endpoints
└── Regional resources
TROUBLESHOOTING
| Issue |
Cause |
Solution |
| High cost |
Over-provisioned |
Right-size, reservations |
| Latency |
Region distance |
Multi-region, CDN |
| Permission denied |
IAM policy |
Review policies |