backend-design-review（后端设计审查）

Backend Design Review

Review Workflow

Follow this systematic review process:

1. Pre-Review Preparation

• Gather design documentation (architecture diagrams, API specs, database schemas, ADRs)
• Understand requirements (functional, non-functional, compliance)
• Define review scope and priorities
• Identify constraints (technology, budget, timeline)

2. API Design Review

• Evaluate RESTful resource modeling, HTTP method usage, status codes
• Review GraphQL schema design, type definitions, query patterns
• Assess gRPC service definitions and protobuf schemas
• Validate API versioning strategy and documentation
• Check authentication, authorization, and security measures

3. Database Design Validation

• Review data modeling, entity relationships, normalization
• Assess schema design, column types, constraints, indexes
• Evaluate query patterns and N+1 query prevention
• Check data integrity rules and referential integrity
• Review scalability approach (sharding, replicas, caching)

4. Architecture Assessment

• Evaluate service boundaries and decomposition
• Review communication patterns (sync/async, event-driven)
• Assess resilience patterns (circuit breakers, retries, timeouts)
• Check service discovery and load balancing design
• Validate data management and consistency strategies

5. Security Review

• Evaluate authentication mechanisms (OAuth 2.0, JWT)
• Review authorization model (RBAC, ABAC)
• Assess data protection (encryption at rest/transit, secrets)
• Check input validation and injection prevention
• Review security monitoring and audit logging

6. Performance & Scalability

• Assess caching strategy (layers, invalidation, TTL)
• Review database indexing and query optimization
• Evaluate horizontal/vertical scaling approach
• Check load balancing and auto-scaling design
• Review asynchronous processing patterns

7. Report Generation

• Categorize findings by severity (Critical, High, Medium, Low)
• Document detailed findings with examples
• Provide specific, actionable recommendations
• Create architecture improvement diagrams
• Define implementation roadmap and priorities

Review Scope

API Design Quality

• RESTful API assessment (resource modeling, HTTP methods, status codes, versioning)
• GraphQL schema review (types, resolvers, complexity, N+1 prevention)
• gRPC service review (protobuf definitions, streaming, error handling)
• API documentation quality (OpenAPI/Swagger completeness)
• API security design (authentication, authorization, rate limiting, validation)

Database Architecture

• Data modeling (entity relationships, normalization, domain alignment)
• Schema design (tables, columns, constraints, indexes, partitioning)
• Query patterns (efficiency, index usage, N+1 prevention)
• Data integrity (referential integrity, constraints, validation)
• Scalability (sharding, read replicas, caching)

Microservices Patterns

• Service boundaries (decomposition, bounded contexts, DDD alignment)
• Communication patterns (sync/async, event-driven, orchestration)
• Data management (database-per-service, eventual consistency, sagas)
• Service discovery (registry, load balancing)
• Resilience (circuit breakers, retries, timeouts, bulkheads)

Integration Architecture

• Integration patterns (API, message queues, event streaming, webhooks)
• Message queue design (selection, schemas, DLQ, idempotency)
• Event streaming (event sourcing, CQRS, stream processing)
• External API integration (retry logic, circuit breakers, versioning)
• Batch processing (ETL, job scheduling, error handling)

Security Architecture

• Authentication design (JWT, OAuth 2.0, session management)
• Authorization design (RBAC, ABAC, permission models)
• Data protection (encryption at rest/transit, secrets management)
• API security (validation, injection prevention, rate limiting)
• Security monitoring (audit logging, anomaly detection)

Severity Levels

Use these severity ratings for findings:

• 🔴 Critical: Security risks, data loss, broken functionality - must fix before implementation
• 🟠 High: Significant flaws affecting scalability, performance, reliability - should fix before go-live
• 🟡 Medium: Moderate issues or best practice deviations - address in next iteration
• 🟢 Low: Minor improvements or optimizations - track for future improvements

Report Structure

Present backend design review findings with:

1. Executive Summary - Project context, review date, overall assessment
2. Review Scope - What was reviewed, depth of review, focus areas
3. Key Findings Summary - Critical and high severity issues overview
4. Detailed Findings - Each finding with severity, description, impact, recommendations, examples
5. Positive Observations - Strengths and good design decisions
6. Recommendations - Prioritized improvements with implementation guidance
7. Architecture Diagrams - Current state and proposed improvements
8. Action Items - Specific tasks with owners, deadlines, and status tracking
9. Next Steps - Immediate actions, short-term tasks, follow-up review schedule

Reference Files

Load detailed guidance based on specific review needs:

• Review Process: See backend-design-review-process.md for comprehensive step-by-step review workflow covering API design, database validation, microservices assessment, security review, and performance evaluation with detailed checklists
• API Design Patterns: See api-design-patterns.md when reviewing RESTful APIs, GraphQL schemas, or gRPC services - includes resource modeling, HTTP methods, status codes, versioning strategies, authentication patterns, and common anti-patterns
• Database Design Patterns: See database-design-patterns.md for detailed guidance on data modeling, normalization, indexing strategies, query optimization, database scaling patterns, NoSQL patterns, and caching strategies
• Microservices & Integration Patterns: See microservices-integration-patterns.md when reviewing microservices architecture, service boundaries, communication patterns, resilience patterns, message queues, event streaming, and distributed system designs
• Report Template: See report-template.md for complete report structure with sections for executive summary, findings, recommendations, architecture diagrams, and action items
• Severity Levels: See severity-levels.md for detailed severity rating criteria (Critical, High, Medium, Low) with examples and action requirements