What are the key metrics to measure in performance testing?

Key Performance Metrics

1. Response Time:

• Time taken to receive response after sending request
• Measured in milliseconds or seconds
• Target: < 2 seconds for web pages, < 500ms for APIs

2. Throughput:

• Number of requests processed per unit time
• Measured in requests/second or transactions/second
• Higher throughput = better performance

3. Concurrent Users:

• Number of users accessing system simultaneously
• Tests system capacity
• Example: 1000 concurrent users

4. Error Rate:

• Percentage of failed requests
• Target: < 1% error rate
• Includes HTTP errors (500, 503, etc.)

5. CPU Utilization:

• Percentage of CPU being used
• Target: < 70-80% under normal load
• High CPU may indicate performance bottleneck

6. Memory Usage:

• Amount of RAM consumed
• Monitor for memory leaks
• Should remain stable over time

7. Network Bandwidth:

• Data transfer rate
• Measured in Mbps or Gbps
• Important for data-heavy applications

8. Database Performance:

• Query execution time
• Connection pool usage
• Database locks and deadlocks

Performance Testing Types

Load Testing:

• Test system under expected load
• Example: 500 concurrent users

Stress Testing:

• Test system beyond normal capacity
• Find breaking point

Spike Testing:

• Sudden increase in load
• Example: Black Friday traffic

Endurance Testing:

• Test system over extended period
• Detect memory leaks

Performance Testing Tools

• JMeter: Open-source load testing tool
• LoadRunner: Enterprise performance testing
• Gatling: Modern load testing framework
• K6: Developer-centric load testing
• Locust: Python-based load testing

Acceptable Performance Benchmarks

• Page load time: < 3 seconds
• API response time: < 500ms
• Database query: < 100ms
• Error rate: < 1%
• CPU usage: < 80%