Convert Figma logo to code with AI

martinus logonanobench

Simple, fast, accurate single-header microbenchmarking functionality for C++11/14/17/20

1,406
81
1,406
17
View on GitHub

Top Related Projects

google's avatar

benchmark

8,962

A microbenchmark support library

google's avatar

googletest

34,586

GoogleTest - Google Testing and Mocking Framework

Quick Overview

Nanobench is a simple, fast, and accurate microbenchmarking library for C++. It provides a straightforward way to measure and compare the performance of small code snippets, functions, or algorithms with minimal overhead and high precision.

Pros

  • Easy to use with a simple API
  • Highly accurate measurements with low overhead
  • Supports multiple output formats (console, JSON, CSV)
  • Header-only library, easy to integrate into existing projects

Cons

  • Limited to C++ projects
  • May require some setup for more complex benchmarking scenarios
  • Not as feature-rich as some larger benchmarking frameworks
  • Documentation could be more comprehensive

Code Examples

  1. Basic benchmark:
#include <nanobench.h>

int main() {
    ankerl::nanobench::Bench().run("example", []() {
        // Code to benchmark
        volatile int x = 1 + 1;
    });
}
  1. Comparing multiple functions:
#include <nanobench.h>
#include <vector>
#include <algorithm>

int main() {
    std::vector<int> v(1000);
    
    ankerl::nanobench::Bench()
        .run("std::sort", [&] {
            std::sort(v.begin(), v.end());
        })
        .run("std::stable_sort", [&] {
            std::stable_sort(v.begin(), v.end());
        });
}
  1. Custom output:
#include <nanobench.h>

int main() {
    ankerl::nanobench::Bench()
        .output(nullptr) // Disable console output
        .run("example", [] {
            volatile int x = 1 + 1;
        })
        .render(ankerl::nanobench::templates::json(), std::cout);
}

Getting Started

To use Nanobench in your C++ project:

  1. Download the nanobench.h header file from the GitHub repository.
  2. Include the header in your C++ file: 
    #include "nanobench.h"
  3. Create a benchmark: 
    int main() {
    ankerl::nanobench::Bench().run("my_benchmark", [] {
        // Your code to benchmark here
    });
    return 0;
}
  4. Compile and run your program to see the benchmark results.

Competitor Comparisons

google logo

benchmark

8,962

A microbenchmark support library

Pros of benchmark

  • More comprehensive feature set, including multithreaded benchmarking
  • Wider adoption and community support
  • Extensive documentation and examples

Cons of benchmark

  • Larger and more complex, potentially overkill for simple use cases
  • Requires more setup and configuration

Code Comparison

nanobench:

#include <nanobench.h>

ankerl::nanobench::Bench().run("my_function", [&] {
    my_function();
});

benchmark:

#include <benchmark/benchmark.h>

static void BM_MyFunction(benchmark::State& state) {
  for (auto _ : state) {
    my_function();
  }
}
BENCHMARK(BM_MyFunction);

Key Differences

  • nanobench is header-only and easier to integrate
  • benchmark offers more granular control over benchmark parameters
  • nanobench provides built-in statistical analysis
  • benchmark has better support for custom output formats

Use Cases

  • nanobench: Quick, simple benchmarks with minimal setup
  • benchmark: Complex benchmarking scenarios, especially in larger projects

Performance

Both libraries have low overhead, but nanobench may have a slight edge in simplicity and compile-time performance.

google logo

googletest

34,586

GoogleTest - Google Testing and Mocking Framework

Pros of Google Test

  • Comprehensive testing framework with a wide range of features
  • Extensive documentation and community support
  • Supports various testing styles (e.g., fixture, parameterized, typed tests)

Cons of Google Test

  • Larger codebase and potential overhead for simple projects
  • Steeper learning curve for beginners
  • May require more setup and configuration

Code Comparison

Google Test:

TEST(FactorialTest, HandlesZeroInput) {
  EXPECT_EQ(Factorial(0), 1);
}

TEST(FactorialTest, HandlesPositiveInput) {
  EXPECT_EQ(Factorial(1), 1);
  EXPECT_EQ(Factorial(2), 2);
  EXPECT_EQ(Factorial(3), 6);
}

Nanobench:

ankerl::nanobench::Bench().run("factorial", [&] {
    ankerl::nanobench::doNotOptimizeAway(Factorial(10));
});

Key Differences

  • Google Test focuses on unit testing and assertions
  • Nanobench specializes in micro-benchmarking and performance measurement
  • Google Test provides a more extensive testing ecosystem
  • Nanobench offers a simpler API for quick benchmarking tasks

Use Cases

  • Choose Google Test for comprehensive unit testing and test-driven development
  • Opt for Nanobench when primarily concerned with performance benchmarking and optimization

Convert Figma logo designs to code with AI

Visual Copilot

Introducing Visual Copilot: A new AI model to turn Figma designs to high quality code using your components.

Try Visual Copilot

README

ankerl::nanobench

ankerl::nanobench logo

Release GitHub license Travis CI Build Status Appveyor Build Status Join the chat at https://gitter.im/nanobench/community

ankerl::nanobench is a platform independent microbenchmarking library for C++11/14/17/20.

#define ANKERL_NANOBENCH_IMPLEMENT
#include <nanobench.h>

int main() {
    double d = 1.0;
    ankerl::nanobench::Bench().run("some double ops", [&] {
        d += 1.0 / d;
        if (d > 5.0) {
            d -= 5.0;
        }
        ankerl::nanobench::doNotOptimizeAway(d);
    });
}

The whole executable runs for ~60ms and prints

|               ns/op |                op/s |    err% |          ins/op |          cyc/op |    IPC |         bra/op |   miss% |     total | benchmark
|--------------------:|--------------------:|--------:|----------------:|----------------:|-------:|---------------:|--------:|----------:|:----------
|                7.52 |      132,948,239.79 |    1.1% |            6.65 |           24.07 |  0.276 |           1.00 |    8.9% |      0.00 | `some double ops`

Which github renders as

ns/opop/serr%ins/opcyc/opIPCbra/opmiss%totalbenchmark
7.52132,948,239.791.1%6.6524.070.2761.008.9%0.00some double ops

The benchmarked code takes 7.52 nanoseconds to run, so ~133 million times per seconds. Measurements fluctuate by 1.1%. On average 6.65 instructions are executed in 24.07 CPU cycles, resulting in 0.276 instructions per cycle. A single branch is in the code, which branch prediction missed in 8.9% of the cases. Total runtime of the benchmark with the name some double ops is 0.00, so just a few milliseconds.

Design Goals

Documentation

Extensive documentation is available.

More

  • Code of Conduct - Contributor Covenant Code of Conduct
  • I need a better logo. Currently I use a small bench. Nanobench. Ha ha.