magnilucaTaskParallelismControl 0.0.1-alpha.1

TaskParallelismControl

.NET library providing extension methods for executing asynchronous operations in parallel with a controlled degree of parallelism (throttling). Useful to avoid saturating network/CPU/resources when processing large collections.

Features

• Controlled concurrency via a fixed number of worker tasks
• Two overloads:
  • ExecuteInParallel<T>(Func<T, Task>, ...)
  • ExecuteInParallel<T, TResult>(Func<T, Task<TResult>>, ...)
• Thread-safe implementation based on ConcurrentQueue<T> and ConcurrentBag<TResult>

Installation

If published on NuGet:

dotnet add package TaskParallelismControl

Local feed scenario (see LocalNuget folder in the solution):

dotnet nuget add source <PATH_TO_LocalNuget> -n LocalNuget
dotnet add package TaskParallelismControl --source LocalNuget

Quick start

Example: with results

using TaskParallelismControl;

var urls = new[]
{
    "https://learn.microsoft.com/",
    "https://learn.microsoft.com/dotnet/",
};

const int maxParallelism = 5;

var results = await urls.ExecuteInParallel(
    async url =>
    {
        using var client = new HttpClient();
        var bytes = await client.GetByteArrayAsync(url);
        return bytes.Length;
    },
    maxParallelism);

Console.WriteLine($"Total bytes: {results.Sum()}");

Example: without results

using TaskParallelismControl;

var items = Enumerable.Range(1, 100);

await items.ExecuteInParallel(
    async i =>
    {
        await Task.Delay(50);
        Console.WriteLine(i);
    },
    degreeOfParallelism: 8);

Notes

• Target framework: net9.0.
• Pack with dotnet pack.

TaskParallelismControl

.NET library providing extension methods for executing asynchronous operations in parallel with a controlled degree of parallelism (throttling). Useful to avoid saturating network/CPU/resources when processing large collections.

Features

• Controlled concurrency via a fixed number of worker tasks
• Two overloads:
  • ExecuteInParallel<T>(Func<T, Task>, ...)
  • ExecuteInParallel<T, TResult>(Func<T, Task<TResult>>, ...)
• Thread-safe implementation based on ConcurrentQueue<T> and ConcurrentBag<TResult>

Installation

If published on NuGet:

dotnet add package TaskParallelismControl

Local feed scenario (see LocalNuget folder in the solution):

dotnet nuget add source <PATH_TO_LocalNuget> -n LocalNuget
dotnet add package TaskParallelismControl --source LocalNuget

Quick start

Example: with results

using TaskParallelismControl;

var urls = new[]
{
    "https://learn.microsoft.com/",
    "https://learn.microsoft.com/dotnet/",
};

const int maxParallelism = 5;

var results = await urls.ExecuteInParallel(
    async url =>
    {
        using var client = new HttpClient();
        var bytes = await client.GetByteArrayAsync(url);
        return bytes.Length;
    },
    maxParallelism);

Console.WriteLine($"Total bytes: {results.Sum()}");

Example: without results

using TaskParallelismControl;

var items = Enumerable.Range(1, 100);

await items.ExecuteInParallel(
    async i =>
    {
        await Task.Delay(50);
        Console.WriteLine(i);
    },
    degreeOfParallelism: 8);

Notes

• Target framework: net9.0.
• Pack with dotnet pack.