Skip to content

perf-test-types.md

Latest commit

 

History

History
219 lines (175 loc) · 5.14 KB

perf-test-types.md

File metadata and controls

219 lines (175 loc) · 5.14 KB

Performance Test Types Guide

Most Common Performance Test Types

1. Load Test

The most fundamental performance test type that verifies system behavior under expected normal conditions.

Purpose:

  • Validate system performance under expected production load
  • Verify system meets performance SLAs
  • Baseline performance metrics

Key Characteristics:

  • Constant number of virtual users (VUs)
  • Steady, predictable load
  • Usually runs for 5-30 minutes

Example Configuration:

export const options = {
  vus: 50,
  duration: '15m',
  thresholds: {
    http_req_duration: ['p95<500'], // 95% of requests within 500ms
    http_req_failed: ['rate<0.01'],  // Less than 1% error rate
  },
};

When to Use:

  • Regular performance validation
  • Baseline performance measurements
  • Pre-production verification

2. Stress Test

Gradually increases load until the system breaks or degrades significantly.

Purpose:

  • Find system breaking points
  • Determine maximum capacity
  • Identify performance bottlenecks
  • Verify system recovery

Key Characteristics:

  • Gradually increasing load
  • Runs until failure or performance degradation
  • Monitors system recovery

Example Configuration:

export const options = {
  stages: [
    { duration: '2m', target: 100 },  // Normal load
    { duration: '5m', target: 500 },  // Ramp up
    { duration: '2m', target: 1000 }, // Breaking point
    { duration: '1m', target: 0 },    // Recovery
  ],
  thresholds: {
    http_req_failed: ['rate<0.05'],
  },
};

When to Use:

  • Capacity planning
  • System limit identification
  • Scaling decisions
  • Failover testing

3. Spike Test

Tests system's ability to handle sudden, large increases in load.

Purpose:

  • Verify system behavior during traffic spikes
  • Test auto-scaling effectiveness
  • Validate system recovery
  • Identify performance under sudden load

Key Characteristics:

  • Sudden increase in load
  • Short duration at peak
  • Monitors recovery time

Example Configuration:

export const options = {
  stages: [
    { duration: '1m', target: 10 },   // Baseline
    { duration: '1m', target: 500 },  // Spike
    { duration: '3m', target: 500 },  // Sustained spike
    { duration: '1m', target: 10 },   // Recovery
  ],
};

When to Use:

  • Sale events (Black Friday)
  • Event-triggered traffic spikes
  • Marketing campaign launches
  • Emergency scenario planning

4. Ramp-up Test (Scalability Test)

Gradually increases load to verify system scaling capabilities.

Purpose:

  • Validate auto-scaling configurations
  • Test gradual system scaling
  • Verify resource allocation
  • Monitor performance during scaling

Key Characteristics:

  • Gradual load increase
  • Step-wise progression
  • Monitoring of scaling events

Example Configuration:

export const options = {
  stages: [
    { duration: '3m', target: 10 },
    { duration: '5m', target: 50 },
    { duration: '7m', target: 100 },
    { duration: '2m', target: 0 },
  ],
};

When to Use:

  • Auto-scaling validation
  • Cloud deployment testing
  • Resource allocation testing
  • Gradual system warm-up

5. Soak Test (Endurance Test)

Long-running test to identify performance degradation over time.

Purpose:

  • Find memory leaks
  • Identify resource depletion
  • Verify system stability
  • Test data handling over time

Key Characteristics:

  • Extended duration (hours/days)
  • Consistent moderate load
  • Resource monitoring

Example Configuration:

export const options = {
  vus: 50,
  duration: '6h',
  thresholds: {
    http_req_duration: ['p95<1000'],
    http_req_failed: ['rate<0.01'],
  },
};

When to Use:

  • Pre-release validation
  • Memory leak detection
  • Database performance testing
  • Long-term stability verification

Other Performance Test Types (Brief Overview)

  1. Breakpoint Test - Identifies exact point where system fails under load

  2. Capacity Test - Determines maximum user load while maintaining SLAs

  3. Volume Test - Tests system with large amounts of data

  4. Isolation Test - Verifies performance of system components in isolation

  5. Configuration Test - Tests performance across different system configurations

  6. Failover Test - Validates system behavior during component failures

  7. Compliance Test - Ensures performance meets regulatory requirements

  8. Baseline Test - Establishes reference point for future comparisons

  9. Smoke Test - Quick test to verify basic performance functionality

  10. Recovery Test - Tests system recovery after failures under load

Best Practices

  1. Test Environment:

    • Use production-like environment
    • Clear test data between runs
    • Monitor system resources

  2. Test Data:

    • Use realistic data sets
    • Maintain data consistency
    • Clean up test data

  3. Monitoring:

    • Track all relevant metrics
    • Monitor system resources
    • Log unusual behavior

  4. Thresholds:

    • Set realistic SLAs
    • Monitor error rates
    • Track response times

Common Metrics to Monitor

  • Response Time (p95, p99)
  • Error Rate
  • Throughput (RPS)
  • CPU Usage
  • Memory Usage
  • Network I/O
  • Database Connections
  • Cache Hit Ratio