DRN.Test.Performance

Name: test-performance
Rating: 76
Author: duranserkan

Performance testing with BenchmarkDotNet and K6 load testing.

When to Apply

•Benchmarking code performance
•Load testing APIs
•Measuring performance regressions
•Generating performance reports

Project Structure

code

DRN.Test.Performance/
├── Benchmark/          # BenchmarkDotNet benchmarks
│   ├── Framework/      # Framework benchmarks
│   └── Sample/         # Application benchmarks
├── K6/                 # K6 load test scripts
│   ├── scripts/        # Test scripts
│   └── config/         # K6 configuration
├── Reports/            # Generated reports
│   ├── BenchmarkDotNet.Artifacts/
│   └── K6Results/
├── Sketch.cs           # Experimental benchmarks
├── Todo.cs             # Planned benchmarks
└── xunit.runner.json

BenchmarkDotNet

Basic Benchmark

csharp

[MemoryDiagnoser]
public class MyBenchmark
{
    private readonly List<int> _data = Enumerable.Range(0, 1000).ToList();
    
    [Benchmark(Baseline = true)]
    public int ForLoop()
    {
        var sum = 0;
        for (var i = 0; i < _data.Count; i++)
            sum += _data[i];
        return sum;
    }
    
    [Benchmark]
    public int LinqSum() => _data.Sum();
    
    [Benchmark]
    public int Foreach()
    {
        var sum = 0;
        foreach (var item in _data)
            sum += item;
        return sum;
    }
}

Running Benchmarks

bash

# Run all benchmarks
dotnet run -c Release --project DRN.Test.Performance

# Run specific benchmark
dotnet run -c Release --project DRN.Test.Performance -- --filter "*MyBenchmark*"

Benchmark Attributes

Attribute	Purpose
`[Benchmark]`	Mark method as benchmark
`[Benchmark(Baseline = true)]`	Set baseline for comparison
`[MemoryDiagnoser]`	Include memory allocation metrics
`[GlobalSetup]`	One-time setup before all iterations
`[IterationSetup]`	Setup before each iteration
`[Params(1, 10, 100)]`	Parameterized benchmarks

K6 Load Testing

Basic K6 Script

javascript

// k6/scripts/api-load-test.js
import http from 'k6/http';
import { check, sleep } from 'k6';

export const options = {
    vus: 10,           // Virtual users
    duration: '30s',   // Test duration
    thresholds: {
        http_req_failed: ['rate<0.01'],       // <1% failures
        http_req_duration: ['p(95)<500'],     // 95th percentile under 500ms
    },
};

export default function () {
    const res = http.get('http://localhost:5988/api/status');
    
    check(res, {
        'is status 200': (r) => r.status === 200,
        'response time < 200ms': (r) => r.timings.duration < 200,
    });
    
    sleep(1);
}

Running K6

bash

# Run load test
k6 run k6/scripts/api-load-test.js

# Run with custom VUs and duration
k6 run --vus 50 --duration 1m k6/scripts/api-load-test.js

# Output to JSON
k6 run --out json=results.json k6/scripts/api-load-test.js

K6 Test Types

Type	Purpose	Configuration
Load Test	Normal load	`vus: 10, duration: '5m'`
Stress Test	Beyond capacity	Ramp up VUs until failure
Spike Test	Sudden load	Quick ramp to high VUs
Soak Test	Extended duration	Low VUs, long duration

Reports

BenchmarkDotNet Output

Reports generated in Reports/BenchmarkDotNet.Artifacts/:

•HTML reports
•Markdown reports
•JSON results

K6 Output

bash

# HTML report (with k6-reporter)
k6 run --out json=report.json script.js

# Grafana + InfluxDB (real-time)
k6 run --out influxdb=http://localhost:8086/k6 script.js

xunit.runner.json

json

{
  "parallelizeAssembly": false,
  "parallelizeTestCollections": false
}

Performance tests should not run in parallel.

Best Practices

Practice	Reason
Run in Release mode	Accurate performance numbers
Warm-up iterations	JIT compilation effects
Multiple iterations	Statistical significance
Isolate tests	No parallel execution
Document baseline	Track regressions

Related Skills

•overview-drn-testing.md - Testing philosophy
•drn-testing.md - Framework.Testing