Assessment Of Memory Consumption While Launching Million Tasks

published the results of testing memory consumption when performing a code that creates a million parallel to the performed consequences. Testing was carried out for a standard program implemented in programming languages ​​Rust, C#, Go, Java, Python and JavaScript.

In Rust, the tasks were created using the framework tokio and based on the library async_std , in C# – API LIST (), JavaScript – async/AWAIT in Node.js, Python – Asyncio, GO – Goroutine, Java – Magrams. Used versions and configuration: Rust 1.82, .NET 9.0.100 with NATIVEAT,
Go 1.23.3, Openjdk/Graalvm 23.0.1, Nodejs 23.2.0, Python 3.13.0.

Results:

  • Starting one task for evaluating memory consumption in Runtime.

    < /center>
  • Launch 10 thousand tasks.

  • Launch of 100 thousand tasks.

  • Launch of a million tasks.

  • Similar last year’s test with the launch of a million tasks.

/Reports, release notes, official announcements.