Tools

This directory contains development and benchmarking tools for the WPF HexEditor project.

📊 ByteProviderBench

Purpose: Performance benchmarking tool for the ByteProvider class.

Description: ByteProviderBench is a benchmarking application that measures the performance of various ByteProvider operations to identify bottlenecks and optimize critical code paths.

Benchmarked Operations

• Read Performance
  • Single byte reads (GetByte())
  • Bulk byte reads (GetBytes())
  • Sequential vs random access patterns
• Write Performance
  • Single byte modifications
  • Bulk paste operations
  • Insert vs overwrite modes
• Search Performance
  • FindFirst() on various data sizes
  • FindAll() performance
  • Cache hit vs cache miss scenarios
• Undo/Redo Performance
  • Stack operations
  • Memory usage during large undo stacks
  • Redo operation speed

Running Benchmarks

cd ByteProviderBench
dotnet run --configuration Release

Note: Always run benchmarks in Release mode for accurate results.

Interpreting Results

The tool outputs:

• Mean - Average execution time
• StdDev - Standard deviation (lower is more consistent)
• Median - Middle value (50th percentile)
• Min/Max - Fastest and slowest runs
• Ops/sec - Operations per second (higher is better)

Use Cases

• Before optimization - Establish baseline performance
• After changes - Verify improvements
• Regression testing - Ensure changes don’t degrade performance
• Platform comparison - Compare .NET Framework vs .NET Core performance

Performance Targets

Based on typical hardware (SSD, 16GB RAM, modern CPU):

Operation	Target	Acceptable
GetByte()	< 1 μs	< 5 μs
FindFirst() 1MB	< 10 ms	< 50 ms
Paste 1KB	< 1 ms	< 10 ms
Undo/Redo	< 100 μs	< 1 ms

Adding New Benchmarks

To add a new benchmark:

1. Create a new benchmark class:

   [MemoryDiagnoser]
   public class MyBenchmark
   {
    private ByteProvider _provider;
   
    [GlobalSetup]
    public void Setup()
    {
        // Initialize
    }
   
    [Benchmark]
    public void MyOperation()
    {
        // Code to benchmark
    }
   }
   

2. Add to program entry point
3. Run and analyze results

Tips for Accurate Benchmarks

• Close all other applications
• Disable antivirus temporarily
• Use consistent test data
• Run multiple iterations (minimum 10)
• Warm up the JIT compiler
• Test on representative data sizes

Integration with CI/CD

ByteProviderBench can be integrated into CI/CD pipelines to:

• Detect performance regressions
• Generate performance reports
• Compare branches
• Track performance trends over time

Example GitHub Actions workflow:

- name: Run Benchmarks
  run: |
    cd Sources/Tools/ByteProviderBench
    dotnet run -c Release --exporters json

- name: Upload Results
  uses: actions/upload-artifact@v2
  with:
    name: benchmark-results
    path: BenchmarkDotNet.Artifacts/

---

🔍 Future Tools

Potential tools to add:

• MemoryProfiler - Track memory usage patterns
• StressTest - Test with extreme file sizes
• FuzzTester - Random input testing for robustness
• UIBenchmark - Measure UI rendering performance
• ComparisonTool - Compare before/after performance

---

📚 Related Documentation

• Core Components - ByteProvider implementation details
• Services Architecture - Service layer performance
• Main README - Project overview

---

✨ Tools by Derek Tremblay and contributors