What this book covers

Chapter 1, Traditional Concurrency, covers common problems with multithreading and solutions to these problems. You will refresh your knowledge about basic locking techniques and how to make locking more efficient.

Chapter 2, Lock-Free Concurrency, goes further into performance optimization. It covers various ways to write concurrent programs without locking, making the code fast and reliable.

Chapter 3, Understanding Parallelism Granularity, explains another important aspect of organizing your parallel code—splitting a computational workload between threads. It introduces coarse-grained and fine-grained approaches, showing their pros and cons.

Chapter 4, Task Parallel Library in Depth, goes into the details of Task Parallel Library—a framework to organize your concurrent program as a set of related tasks. You will find the internals of TPL reviewed and explained.

Chapter 5, C# Language Support for Asynchrony, is a deep dive into the C# language infrastructure. The chapter shows exactly how the async and await keywords work and how you can write your own await-compatible code.

Chapter 6, Using Concurrent Data Structures, covers the use of data structures in a concurrent program in detail, including standard .NET concurrent collections and custom thread safe collections implementations.

Chapter 7, Leveraging Parallel Patterns, reviews programming patterns related to parallel applications. The chapter describes different kinds of patterns—historical .NET idioms, useful code snippets, and a high-level parallel pipeline pattern.

Chapter 8, Server-Side Asynchrony, is a solution description to the problem of using asynchrony on the server. It explains why it is very important to distinguish asynchrony from parallelism, and how it can affect the scalability and reliability of your server.

Chapter 9, Concurrency in the User Interface, describes the details of how the user interface is implemented, what a message loop is, and why it is very important to keep the UI thread nonblocked.

Chapter 10, Troubleshooting Parallel Programs, explains how to find out what is wrong with your parallel program. You will learn how to write unit tests for an asynchronous code, how to debug it, and find performance bottlenecks.