Multi Cloud Architecture

1---
2name: multi-cloud-architecture
3description: Design multi-cloud architectures using a decision framework to select and integrate services across AWS, Azure, and GCP. Use when building multi-cloud systems, avoiding vendor lock-in, or leveraging best-of-breed services from multiple providers.
4---
5 
6# Multi-Cloud Architecture
7 
8Decision framework and patterns for architecting applications across AWS, Azure, and GCP.
9 
10## Do not use this skill when
11 
12- The task is unrelated to multi-cloud architecture
13- You need a different domain or tool outside this scope
14 
15## Instructions
16 
17- Clarify goals, constraints, and required inputs.
18- Apply relevant best practices and validate outcomes.
19- Provide actionable steps and verification.
20- If detailed examples are required, open `resources/implementation-playbook.md`.
21 
22## Purpose
23 
24Design cloud-agnostic architectures and make informed decisions about service selection across cloud providers.
25 
26## Use this skill when
27 
28- Design multi-cloud strategies
29- Migrate between cloud providers
30- Select cloud services for specific workloads
31- Implement cloud-agnostic architectures
32- Optimize costs across providers
33 
34## Cloud Service Comparison
35 
36### Compute Services
37 
38| AWS | Azure | GCP | Use Case |
39|-----|-------|-----|----------|
40| EC2 | Virtual Machines | Compute Engine | IaaS VMs |
41| ECS | Container Instances | Cloud Run | Containers |
42| EKS | AKS | GKE | Kubernetes |
43| Lambda | Functions | Cloud Functions | Serverless |
44| Fargate | Container Apps | Cloud Run | Managed containers |
45 
46### Storage Services
47 
48| AWS | Azure | GCP | Use Case |
49|-----|-------|-----|----------|
50| S3 | Blob Storage | Cloud Storage | Object storage |
51| EBS | Managed Disks | Persistent Disk | Block storage |
52| EFS | Azure Files | Filestore | File storage |
53| Glacier | Archive Storage | Archive Storage | Cold storage |
54 
55### Database Services
56 
57| AWS | Azure | GCP | Use Case |
58|-----|-------|-----|----------|
59| RDS | SQL Database | Cloud SQL | Managed SQL |
60| DynamoDB | Cosmos DB | Firestore | NoSQL |
61| Aurora | PostgreSQL/MySQL | Cloud Spanner | Distributed SQL |
62| ElastiCache | Cache for Redis | Memorystore | Caching |
63 
64**Reference:** See `references/service-comparison.md` for complete comparison
65 
66## Multi-Cloud Patterns
67 
68### Pattern 1: Single Provider with DR
69 
70- Primary workload in one cloud
71- Disaster recovery in another
72- Database replication across clouds
73- Automated failover
74 
75### Pattern 2: Best-of-Breed
76 
77- Use best service from each provider
78- AI/ML on GCP
79- Enterprise apps on Azure
80- General compute on AWS
81 
82### Pattern 3: Geographic Distribution
83 
84- Serve users from nearest cloud region
85- Data sovereignty compliance
86- Global load balancing
87- Regional failover
88 
89### Pattern 4: Cloud-Agnostic Abstraction
90 
91- Kubernetes for compute
92- PostgreSQL for database
93- S3-compatible storage (MinIO)
94- Open source tools
95 
96## Cloud-Agnostic Architecture
97 
98### Use Cloud-Native Alternatives
99 
100- **Compute:** Kubernetes (EKS/AKS/GKE)
101- **Database:** PostgreSQL/MySQL (RDS/SQL Database/Cloud SQL)
102- **Message Queue:** Apache Kafka (MSK/Event Hubs/Confluent)
103- **Cache:** Redis (ElastiCache/Azure Cache/Memorystore)
104- **Object Storage:** S3-compatible API
105- **Monitoring:** Prometheus/Grafana
106- **Service Mesh:** Istio/Linkerd
107 
108### Abstraction Layers
109 
110```
111Application Layer
112    ↓
113Infrastructure Abstraction (Terraform)
114    ↓
115Cloud Provider APIs
116    ↓
117AWS / Azure / GCP
118```
119 
120## Cost Comparison
121 
122### Compute Pricing Factors
123 
124- **AWS:** On-demand, Reserved, Spot, Savings Plans
125- **Azure:** Pay-as-you-go, Reserved, Spot
126- **GCP:** On-demand, Committed use, Preemptible
127 
128### Cost Optimization Strategies
129 
1301. Use reserved/committed capacity (30-70% savings)
1312. Leverage spot/preemptible instances
1323. Right-size resources
1334. Use serverless for variable workloads
1345. Optimize data transfer costs
1356. Implement lifecycle policies
1367. Use cost allocation tags
1378. Monitor with cloud cost tools
138 
139**Reference:** See `references/multi-cloud-patterns.md`
140 
141## Migration Strategy
142 
143### Phase 1: Assessment
144- Inventory current infrastructure
145- Identify dependencies
146- Assess cloud compatibility
147- Estimate costs
148 
149### Phase 2: Pilot
150- Select pilot workload
151- Implement in target cloud
152- Test thoroughly
153- Document learnings
154 
155### Phase 3: Migration
156- Migrate workloads incrementally
157- Maintain dual-run period
158- Monitor performance
159- Validate functionality
160 
161### Phase 4: Optimization
162- Right-size resources
163- Implement cloud-native services
164- Optimize costs
165- Enhance security
166 
167## Best Practices
168 
1691. **Use infrastructure as code** (Terraform/OpenTofu)
1702. **Implement CI/CD pipelines** for deployments
1713. **Design for failure** across clouds
1724. **Use managed services** when possible
1735. **Implement comprehensive monitoring**
1746. **Automate cost optimization**
1757. **Follow security best practices**
1768. **Document cloud-specific configurations**
1779. **Test disaster recovery** procedures
17810. **Train teams** on multiple clouds
179 
180## Reference Files
181 
182- `references/service-comparison.md` - Complete service comparison
183- `references/multi-cloud-patterns.md` - Architecture patterns
184 
185## Related Skills
186 
187- `terraform-module-library` - For IaC implementation
188- `cost-optimization` - For cost management
189- `hybrid-cloud-networking` - For connectivity
190