2.1 Performance Testing

1. What Is Performance Testing

• How the system operates under different loads.

2. Performance Testing Metrics

2.1. Throughput

• Throughput: the number of client requests processed per unit of time.
• Throughput ≈ concurrency / average response time (when the system is in a steady state and the definitions are consistent).

2.1.1. TPS

• Transactions Per Second: the number of transactions per second.

2.1.2. QPS

• Query Per Second: the number of queries per second.

2.2. Concurrency

• Concurrent users: the number of users submitting requests to the system at the same physical moment.
• Online users: the number of users who access the system during a period of time; these users are not necessarily submitting requests at the same time.
• System users: the total number of registered users in the system.

2.3. Response Time

• Response time is the time from when the user initiates a request from the client until the client receives the result returned by the server.
• Response time = network round-trip time + server processing time.

2.4. Resource Utilization

• Resource utilization: how system resources are being used.
• Resource utilization = resource usage / total available resources * 100%.
• Reference utilization values from the original note (not universal thresholds):
  • CPU no higher than 80%.
  • Memory no higher than 80%.
  • Disk no higher than 90%.

2.5. Error Rate

• Error rate: the number of failed requests.
• Error rate = failed transactions / total transactions * 100%.
• Reference error rate from the original note (not a universal threshold):
  • Below 0.5%.
  • Error types depend on the specific system and failure mode.

3. References

• What Is the Difference Between Stress Testing and Performance Testing? - Zhihu