bench_matrix 0.2.0

A utility crate for defining and running parameterized benchmarks, with optional Criterion integration.
Documentation

bench_matrix

Crates.io Docs.rs License: MPL 2.0

bench_matrix is a Rust utility crate that supercharges your parameterized benchmarks. It provides a powerful and ergonomic framework for running benchmarks across a complex matrix of configurations, integrating seamlessly with the Criterion harness.

Stop writing repetitive benchmark functions. Define your parameter axes once, and let bench_matrix handle the rest, generating a full suite of benchmarks with clean, hierarchical reporting.

Why use bench_matrix?

  • Eliminate Boilerplate: Define your parameters (e.g., data sizes, algorithms, concurrency levels) in one place. bench_matrix generates the Cartesian product, ensuring every combination is tested without repetitive code.
  • Memory Efficient: Lazily generates benchmark combinations on the fly. You can define a test matrix with millions of variants without consuming gigabytes of memory upfront.
  • Clean, Hierarchical Reports: Automatically creates well-named Criterion groups, leading to organized and readable benchmark results (e.g., MySuite/Algorithm-QuickSort_DataSize-1000).
  • Seamless Criterion Integration: Built from the ground up to work with Criterion, leveraging its powerful statistical analysis and plotting features.
  • Async & Sync Ready: Provides dedicated, consistent APIs for both synchronous (SyncBenchmarkSuite) and asynchronous (AsyncBenchmarkSuite) code.
  • Type-Safe & Customizable: Use your own strongly-typed configuration structs and hook into a flexible lifecycle with setup, teardown, and global_setup functions.

A Quick Look

Here's how you can set up a benchmark for a function across multiple data sizes and processing intensities:

// In benches/my_bench.rs
use bench_matrix::{criterion_runner::sync_suite::SyncBenchmarkSuite, MatrixCellValue};
use criterion::{criterion_group, criterion_main, Criterion, Throughput};

fn my_benchmark_function(c: &mut Criterion) {
    let parameter_axes = vec![
        // Axis 1: Number of data elements
        vec![MatrixCellValue::Unsigned(100), MatrixCellValue::Unsigned(1000)],
        // Axis 2: Processing intensity
        vec![MatrixCellValue::String("Low".to_string()), MatrixCellValue::String("High".to_string())],
    ];

    let parameter_names = vec!["Elements".to_string(), "Intensity".to_string()];

    // Define your config struct, state, extractor, and lifecycle functions...
    // (See the Usage Guide for full details)

    let suite = SyncBenchmarkSuite::new(
        c, "DataProcessingSuite".to_string(), None, parameter_axes,
        Box::new(my_extractor_fn),
        my_setup_fn,
        my_logic_fn,
        my_teardown_fn,
    )
    .parameter_names(parameter_names)
    .throughput(|cfg: &MyConfig| Throughput::Elements(cfg.data_elements as u64));

    suite.run();
}

criterion_group!(benches, my_benchmark_function);
criterion_main!(benches);

This will produce benchmark results like:

  • DataProcessingSuite/Elements-100_Intensity-Low
  • DataProcessingSuite/Elements-100_Intensity-High
  • DataProcessingSuite/Elements-1000_Intensity-Low
  • DataProcessingSuite/Elements-1000_Intensity-High

Installation

Add bench_matrix and its companions to the [dev-dependencies] section of your Cargo.toml:

[dev-dependencies]
bench_matrix = "0.2.0" # Replace with the latest version
criterion = "0.5"
tokio = { version = "1", features = ["full"] } # Required for async benchmarks

The criterion_integration feature is enabled by default.

Documentation