explorer-monorepo/docs/specs/deployment/scaling.md

Scaling Strategy Specification

Overview

Horizontal and vertical scaling strategies for the platform.

Horizontal Scaling Patterns

Application Scaling

Method: Add more service instances Auto-scaling: Based on CPU, memory, request rate Load Balancing: Distribute traffic across instances

Database Scaling

Read Replicas: Scale read capacity Sharding: Partition data for write scaling Connection Pooling: Efficient connection management

Database Scaling

Read Replicas

Strategy: Multiple read replicas for read-heavy workloads Replication: Async replication from primary Use Case: API read operations

Sharding

Strategy: Partition data by chain_id Implementation: Database partitioning Use Case: Very large datasets

Caching Strategy

Redis Cache

Use Cases:

• API response caching
• Session storage
• Rate limiting counters
• Frequently accessed data

CDN Caching

Use Cases:

• Static assets
• API responses (short TTL)

Load Balancing Configuration

Method: Kubernetes service or external load balancer Algorithm: Round-robin or least connections Health Checks: Regular health checks, route to healthy instances

References

• Infrastructure: See infrastructure.md
• CI/CD: See cicd.md