(Archived) Repository containing benchmarks organized by tech and discussions around it. The benchmarks are not limited to a single programming language, or even to a single framework and are focused on general purpose.
Each benchmark will be composing by the following parts:
- Briefly explanation about the benchmark objective
- Code that was used to gather the benchmark information
- Benchmark results table with data extracted from the code execution
Important
Benchmark methods can vary from benchmark to benchmark and will try to use the most reliable possible tools. Some benchmarks, mainly the specialized ones, are affected by cold starts and other variables so the methods used in them may also vary. We have a curated list of methods and tecniques that can be used to create a benchmark and it is listed below. It's important to note that the list can also vary with future updates.
| Tool | Link |
|---|---|
| hyperfine - a command line benchmarking tool | https://github.com/sharkdp/hyperfine |
| BenchmarkDotNet - powerful .net library for benchmarking | https://github.com/dotnet/BenchmarkDotNet |
| performance.now() - Node.js core library | - |
You can contribute with meaningful benchmarks, fixes and suggestions to see a detailed information about contrbuting, click here.
Tip
Take care about you submissions, meaningful and concise benchmarks are valued and too sparse submissions won't be accepted. Try to explain the objective of the benchmark clearly.
Caution
Benchmarks that depends on libraries, frameworks, external apis or any kind of dependency that is not present in the base runtime won't be accepted.
