← スキル一覧に戻る

cloud-design
by malston
⭐ 1🍴 0📅 2026年1月24日
SKILL.md
name: cloud-design description: Design resilient, scalable multi-cloud architectures with best practices for AWS, Azure, and GCP. Use when creating cloud architecture diagrams, planning cloud infrastructure, or designing cloud migration strategies.
Cloud Architecture Design
Principles and patterns for designing resilient, scalable cloud architectures across AWS, Azure, and GCP.
When to Use
- Designing new cloud architectures
- Planning multi-cloud strategies
- Migrating workloads to the cloud
- Creating high-level architecture diagrams
- Evaluating cloud service selection
Key Design Principles
1. Design for Failure
- Region/Zone Redundancy: Multi-AZ/zone deployments
- Service Isolation: Independent failure domains
- Graceful Degradation: Circuit breakers, fallbacks, timeouts
- Self-Healing: Auto-recovery, health checks, auto scaling
- Disaster Recovery: RPO/RTO-driven backup and recovery
2. Scalability Patterns
- Horizontal Scaling: Stateless services with auto-scaling
- Vertical Partitioning: Services by function
- Horizontal Partitioning: Sharding and data partitioning
- Caching: Multi-level caching strategy
- Asynchronous Processing: Event-driven, queue-based architectures
3. Security by Design
- Defense in Depth: Multiple security controls
- Least Privilege: Minimal required permissions
- Zero Trust: Assume breach mentality
- Data Protection: Encryption at rest and in transit
- Immutable Infrastructure: CI/CD with infrastructure as code
4. Cost Optimization
- Right-sizing: Match resources to workload
- Elasticity: Scale with demand
- Reserved Capacity: Predictable workloads
- Spot/Preemptible: Non-critical workloads
- Serverless First: Pay for execution, not idle
5. Operational Excellence
- Monitoring: Comprehensive metrics, logs, traces
- Automation: Self-service, infrastructure as code
- Standardization: Patterns, templates, naming
- Documentation: Architecture decisions, runbooks
- Observability: Business and technical insights
Multi-Cloud Design Patterns
Service Equivalents Map
| Pattern | AWS | Azure | GCP |
|---|---|---|---|
| Compute | EC2, Lambda | VM, Functions | Compute Engine, Functions |
| Container | ECS, EKS | AKS, Container Apps | GKE, Cloud Run |
| Serverless | Lambda, Step Functions | Functions, Logic Apps | Functions, Workflows |
| Storage | S3, EFS | Blob, Files | GCS, Filestore |
| Database | RDS, DynamoDB | SQL, Cosmos DB | Cloud SQL, Firestore |
| Networking | VPC, Route53 | VNET, DNS | VPC, Cloud DNS |
| CDN | CloudFront | Front Door | Cloud CDN |
| API Gateway | API Gateway | API Management | Apigee, API Gateway |
Multi-Cloud Strategies
-
Cloud-Agnostic
- Abstract cloud services behind interfaces
- Use common tools and patterns (Kubernetes, Terraform)
- Focus on portability over cloud-specific features
-
Best-of-Breed
- Select optimal services from each provider
- Accept higher integration complexity
- Optimize for feature richness and innovation
-
Primary-Secondary
- Main provider for most workloads
- Secondary for specific use cases or DR
- Balance between optimization and complexity
-
Cloud-Specific
- Embrace native services for each workload
- Optimize for cloud provider strengths
- Accept potential lock-in for better performance
Architecture Archetypes
Web Application
[Users] → [CDN] → [Load Balancer] → [Web Tier] → [API Tier] → [Database]
↑ ↑ ↑ ↑
└───────[Caching Layer]───────────┘ |
|
[Authentication] → [Identity Provider] |
|
[Monitoring] → [Logging] → [Alerting] |
↑ |
└─────────────────────────────────────────┘
Event-Driven Microservices
[Event Source] → [Event Hub/Queue] → [Processing Service] → [Database/Storage]
↓ | ↑
[Event Store] ← [Event Sink] ←┘ |
↓ |
[Analytics] |
|
[Monitoring] → [Tracing] → [Logging] |
↑ |
└───────────────────────────────────────────┘
Data Processing Pipeline
[Data Sources] → [Ingestion] → [Storage] → [Processing] → [Serving]
↓ ↓ ↓ ↓
[Monitoring] [Governance] [Orchestration] [Security]
Provider-Specific Best Practices
AWS
- Use multi-AZ deployments with auto-scaling groups
- Implement VPC flow logs and GuardDuty for security
- Use IAM roles for service-to-service communication
- Leverage S3 lifecycle policies for cost management
- Implement AWS Well-Architected Framework principles
Azure
- Deploy across availability zones with VMSS
- Use Azure Security Center and Sentinel
- Implement managed identities for service authentication
- Use Azure Reserved Instances for predictable workloads
- Follow Azure Well-Architected Framework guidelines
GCP
- Design for regional deployments with global load balancing
- Implement VPC Service Controls and Security Command Center
- Use Workload Identity for service-to-service authentication
- Leverage committed use discounts for predictable workloads
- Follow Google Cloud Architecture Framework principles
Additional Resources
- Cloud provider well-architected frameworks
- Reference architectures for common patterns
- Migration tools and methodologies
- Architecture decision record templates
スコア
総合スコア
50/100
リポジトリの品質指標に基づく評価
✓SKILL.md
SKILL.mdファイルが含まれている
+20
○LICENSE
ライセンスが設定されている
0/10
○説明文
100文字以上の説明がある
0/10
○人気
GitHub Stars 100以上
0/15
○最近の活動
3ヶ月以内に更新がある
0/10
○フォーク
10回以上フォークされている
0/5
✓Issue管理
オープンIssueが50未満
+5
✓言語
プログラミング言語が設定されている
+5
○タグ
1つ以上のタグが設定されている
0/5
レビュー
💬
レビュー機能は近日公開予定です