C# 10 in a Nutshell

Chapter 22. Parallel Programming

In this chapter, we cover the multithreading APIs and constructs aimed at leveraging multicore processors:

• Parallel LINQ, or PLINQ

• The Parallel class

• The task parallelism constructs

• The concurrent collections

These constructs are collectively known (loosely) as Parallel Framework (PFX). The Parallel class together with the task parallelism constructs is called the Task Parallel Library (TPL).

You’ll need to be comfortable with the fundamentals in Chapter 14 before reading this chapter—particularly locking, thread safety, and the Task class.

Note

.NET offers a number of additional specialized APIs to help with parallel and asynchronous programming:

• System.Threading.Channels.Channel is a high-performance asynchronous producer/consumer queue, introduced in .NET Core 3.

• Microsoft Dataflow (in the System.Thread⁠ing.Tasks​.Dataflow namespace) is a sophisticated API for creating networks of buffered blocks that execute actions or data transformations in parallel, with a semblance to actor/agent programming.

• Reactive Extensions implements LINQ over IObservable (an alternative abstraction to IAsyncEnumerable) and excels at combining asynchronous streams. Reactive extensions ships in the System.Reactive NuGet package.

Why PFX?

Over the past 15 years, CPU manufacturers have shifted from single-core to multicore processors. This is problematic for us as programmers because single-threaded code does not automatically run faster as a result ...