Capacity Planning
Workflows
- Baseline: Measure current resource usage
- Forecast: Project future growth
- Size: Calculate required resources
- Buffer: Add headroom for spikes
- Monitor: Track actual vs. predicted
Key Metrics
Compute
- CPU utilization (target: 60-70%)
- Memory usage
- Request latency (P50, P95, P99)
Storage
- Disk IOPS
- Throughput (MB/s)
- Capacity growth rate
Network
- Bandwidth utilization
- Connection counts
- Packet loss
Estimation Framework
Little's Law
L = λ × W
L = Average number of items in system
λ = Average arrival rate
W = Average time in system
Example Calculation
Given:
- 1000 requests/second
- 100ms average response time
Required concurrent connections:
L = 1000 × 0.1 = 100 concurrent connections
Resource Sizing
Database Connections
connections = (requests_per_second × avg_query_time) × 1.5
Memory
memory = (concurrent_users × memory_per_user) + base_overhead
CPU Cores
cores = (peak_rps × cpu_time_per_request) / target_utilization
Growth Planning
Traffic Growth
- Historical growth rate
- Planned marketing/launches
- Seasonal patterns
Data Growth
- Records per day
- Record size
- Retention policy
Capacity Planning Document
- Current state metrics
- Growth assumptions
- Resource projections (3, 6, 12 months)
- Cost estimates
- Scaling triggers and thresholds
