Skip to content

benchmark

Directory actions

More options

Directory actions

More options

Latest commit

 

History

History
 
 

benchmark

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
CMakeLists.txt
CMakeLists.txt
 
 
README.md
README.md
 
 
url_parsing_bench.cpp
url_parsing_bench.cpp
 
 

README.md

URL Parsing Benchmarks

Performance benchmarks for the skyr-url library.

Building

Enable benchmarks during CMake configuration:

cmake -B _build -G Ninja -Dskyr_BUILD_BENCHMARKS=ON
cmake --build _build --target url_parsing_bench

Running

Basic benchmark

./_build/benchmark/url_parsing_bench

Default: 10,000 iterations × 34 URLs = 340,000 parses

Custom iteration count

# Run 100,000 iterations (3.4M URLs)
./_build/benchmark/url_parsing_bench 100000

# Quick test with 1,000 iterations
./_build/benchmark/url_parsing_bench 1000

Profiling

macOS (with Xcode Instruments)

Requires Xcode - Install from App Store or https://developer.apple.com/download/

# Verify Xcode tools are installed
xctrace version

# Build with debug symbols
cmake -B _build -G Ninja \
  -Dskyr_BUILD_BENCHMARKS=ON \
  -DCMAKE_BUILD_TYPE=RelWithDebInfo

cmake --build _build --target url_parsing_bench

# Profile with xctrace (modern replacement for 'instruments' command)
xctrace record --template 'Time Profiler' \
  --output /tmp/url_bench.trace \
  --launch ./_build/benchmark/url_parsing_bench 50000

# Open results in Instruments GUI
open /tmp/url_bench.trace

macOS (with sample - no Xcode needed)

Built-in - No installation required

# Build with debug symbols
cmake -B _build -G Ninja \
  -Dskyr_BUILD_BENCHMARKS=ON \
  -DCMAKE_BUILD_TYPE=RelWithDebInfo

cmake --build _build --target url_parsing_bench

# Profile with sample (10 second sample)
sample url_parsing_bench 10 -file /tmp/profile.txt &
./_build/benchmark/url_parsing_bench 50000

# View results
open /tmp/profile.txt

Linux (with perf)

# Build with debug symbols
cmake -B _build -G Ninja \
  -Dskyr_BUILD_BENCHMARKS=ON \
  -DCMAKE_BUILD_TYPE=RelWithDebInfo

cmake --build _build --target url_parsing_bench

# Profile with perf
perf record -g ./_build/benchmark/url_parsing_bench 50000
perf report

# Or generate flamegraph
perf script | stackcollapse-perf.pl | flamegraph.pl > url_parsing.svg

All platforms (with Valgrind callgrind)

# Profile with callgrind
valgrind --tool=callgrind --callgrind-out-file=callgrind.out \
  ./_build/benchmark/url_parsing_bench 1000

# Visualize with kcachegrind (Linux) or qcachegrind (macOS)
qcachegrind callgrind.out

Understanding Results

The benchmark tests 34 diverse URL patterns:

  • Simple ASCII URLs
  • URLs with query parameters and fragments
  • Internationalized domain names (IDN)
  • Unicode in paths
  • Percent-encoded URLs
  • IPv4 and IPv6 addresses
  • Edge cases (data URIs, mailto, etc.)

Typical Performance

On modern hardware (Apple M1/M2, Intel i7/i9, AMD Ryzen):

  • Average: 2-4 µs/URL
  • Throughput: 250,000 - 500,000 URLs/second

Interpreting Results

Good performance:

  • < 5 µs/URL average

  • 200,000 URLs/second throughput

Investigate if:

  • 10 µs/URL average

  • < 100,000 URLs/second throughput

Use profiling to identify hotspots before optimization.

Adding New Test Cases

Edit url_parsing_bench.cpp and add URLs to the test_urls vector:

const std::vector<std::string> test_urls = {
    // Your new test URLs here
    "http://your-test-url.com/path",
    // ...
};

Rebuild and re-run the benchmark.