Understanding Test Statistics

Total Requests #

The total number of HTTP requests sent during the test. This number can be high because each connection continuously sends requests until the test duration ends.

Calculation Example: For example: 30 connections × ~18 requests/second × 30 seconds ≈ 16,200 requests

Use Cases:

– Measure server throughput capacity

– Compare request handling at different load levels

– Identify maximum requests per second (RPS) your server can handle

**Note:** High numbers are normal with fast servers. Each connection sends requests as quickly as the server responds.

Successful Requests #

*Requests that returned HTTP status codes 200-299. These indicate the server successfully processed the request.

Calculation: Success Rate = (Successful / Total) × 100%

Use Cases:

– Monitor server reliability under load

– Detect when server starts failing requests

– Calculate uptime and availability metrics

**Note:** A high success rate (>95%) indicates stable server performance.

Failed Requests #

Requests that returned error status codes (4xx, 5xx) or encountered network errors (timeouts, connection refused, DNS failures).

Calculation: Failed = Total – Successful

Use Cases:

– Identify server stress points

– Monitor error rates under load

– Test error handling and recovery

**Note:** Some failures are expected under heavy load. Monitor the pattern and types of errors.

Average Response Time #

The average time (in milliseconds) from sending a request to receiving a complete response. Lower is better.

Calculation: Average = Sum of all response times / Number of successful requests

Use Cases:

– Measure server performance and speed

– Compare performance before/after optimizations

– Identify performance degradation under load

– Set performance benchmarks

**Note:** Response times typically increase as load increases. Monitor for sudden spikes.

Requests per Second (RPS) #

The average number of requests processed per second during the test. This is a key performance metric.

Calculation: RPS = Total Requests / Test Duration (seconds)

Use Cases:

– Measure server capacity and throughput

– Compare performance across different configurations

– Plan server scaling requirements

– Set performance targets

**Note:** Higher RPS indicates better server capacity. Compare with your expected traffic.